Skip to content
Publicly Available Published by De Gruyter March 6, 2023

43rd annual conference of the association of clinical biochemists in Ireland (ACBI 2021)

Online event

Conference Secretary: Dr Seán Costelloe

Abstract 01

Manufacturer product bulletin: What’s all the phos about?

Bransfield A, RicoRios N & Costelloe SJ

Department of Clinical Biochemistry, Cork University Hospital (CUH), Wilton, Co. Cork, Republic of Ireland.

Background

Automated estimation of lipaemia, haemolysis, and icterus indices (HIL) is a standard component of sample analysis in clinical biochemistry. In August 2021, Beckman Coulter Diagnostics (ordinal scale used for HIL) issued a product notice1 changing the haemolysis index (HI) cut-off for phosphate from HI≥4 to HI≥1. Use of the pre-existing HI cut-off may have led to the inappropriate release of critically low phosphate results (≤0.3 mmol/L) where HI was between 1 and 3. We aimed to assess the effects of this change, the following analyses were performed for samples with HI0-3: a comparison of the median (MDN) phosphate values observed; comparison of the frequency of critically low phosphate values for each HI grade; frequency of phosphate changes that may be considered clinically significant.

Methods

Data for one year were gathered from laboratory databases at CUH. MDN phosphate values for HI0-3 were compared using Kruskal–Wallace one-way ANOVA. For the same datasets, the frequency of critically low hypophosphataemia was compared using chi-square test. Clinically significant changes in phosphate concentration were defined as those breaching a reference change value (RCV) of 1.97%, calculated as √((Analytical coefficient of variation (CV))2 + (intra-individual biological CV)2). The frequency of RCV breaches was compared between HI0-3 using the chi-square test.

Results

Phosphate concentration (MDN ± interquartile range (mmol/L)) increases significantly (P<0.000001) with increasing haemolysis index: HI0=1.11 ± 0.15, n=305,987; HI1=1.16 ± 0.18, n=4,701; HI2=1.19 ± 0.18, n=1,980; HI3=1.23 ± 0.18, n=521. Frequency of critical hypophosphataemia was as follows: HI0=0.019%; HI1=0.021%; HI2=0.051%; HI3=0.000% (P<0.0001). Frequency of RCV breaches was as follows: HI0=0.970%; HI1=0.838%; HI2=0.858%; HI3=0.738% (P<0.0001).

Conclusion

Median phosphate concentration increased with increasing HI. Although the frequency of critical hypophosphataemia and RCV breaches was found to be statistically significantly different between HI levels, this was not considered to be of clinical consequence. Therefore, the historical use of a HI cut-off of 4, rather than 1 did not represent a risk to patient care.

Reference

  • 1. Beckman Coulter IPN-000624, August 2021.

Abstract 02

MMUH Service Review of Simultaneous Lipase and Amylase Requesting

Cullen RE, Lee MJ & Lee GR

Department of Clinical Biochemistry and Diagnostic Endocrinology, Mater Misericordiae University Hospital, Dublin, Ireland (MMUH)

Background

Total amylase and/or lipase values >3 times the upper reference limit (URL) remain central to evaluating pancreatic damage, including the diagnosis of Acute Pancreatitis (AP). Recently updated instrumentation in the MMUH (i.e. Abbott AlinityTM) prompted the evaluation of current practice (2018–2020), to help inform subsequent diagnostic practice/pathways involving either/both biomarker(s) with potential lipase test repatriation.

Methods

We retrospectively reviewed simultaneous lipase and total amylase requests (n=368, Px 119; 63(F)/56(M), age [IQR] 63[47–74]) and assessed test concordance by evaluation against relevant AP diagnostic thresholds (>2 and >3 URL: >291U/L amylase, >234U/L lipase). Together with details of epigastric pain and imaging suggestive of AP, the Atlanta classification1 was used to diagnose AP where 2/3 of the above criteria were fulfilled. We also analytically verified the Sentinel lipase assay on the Abbott AlinityTM against the current referral method (Abbott lipase, Architect c8000 analyser).

Results

Lipase requesting in MMUH increased by 40% in 2020, mainly (89%; 368) (%; n) for protocol driven clinical trials and the remainder (11%; 45) for ruling out AP. In 2020, 38% (n=75Px) of total amylase values >2URL (n=197) were due to confirmed pancreatitis. Total amylase and lipase dual requests demonstrated 64% concordance for evaluating AP. Lipase appeared to be more specific in detecting pancreatic damage. Alinity-based analysis showed acceptable imprecision (CV=2.9%) and compared well with the current method on Passing–Bablok regression [95% CI], slope=0.98 [0.90–1.00], Intercept =1.5 [–0.36 to 3.50].

Conclusion

The analytical performance of the Alinity/Sentinal Lipase assay study supported repatriation of lipase testing with anticipated cost and logistical improvements and more timely patient management. Our results support a dual marker approach with our new biochemical algorithm for evaluating pancreatic damage.

Reference

  • 1. Banks, P. A et al. (2013) Classification of acute pancreatitis-2012: revision of the Atlanta classification and definitions by international consensus’, Gut, 62, 102–111.

Abstract 03

BNP and NT-proBNP Concordance for Ruling Out Heart Failure

Cunning C & Lee G

Department of Clinical Biochemistry and Diagnostic Endocrinology, Mater Misericordiae University Hospital, Dublin, Ireland (MMUH)

Background

The natriuretic peptides B-type natriuretic peptide (BNP) and NT-proBNP have long standing use in the diagnosis and monitoring of heart failure (HF). Several factors, including increased use of Angiotensin Receptor-Neprilysin Inhibitors e.g. Entresto®, which increases BNP but not NTproBNP and increased accessibility of NP testing to primary care, prompted our laboratory to introduce NT-proBNP in place of BNP. Although BNP and NT-proBNP have shown comparable diagnostic and prognostic performance, discordance of up to 28% has been reported in the acute setting. As part of implementing NT-proBNP, we were interested in studying the concordance between BNP and NT-proBNP.

Methods

BNP and NT-proBNP were both analysed prospectively on plasma samples (EDTA, n=1048) received for routine patient management between September 2020 and January 2021. Abbott Architect immunoassay I2000SR and Abbott Alinity ci-series analysers were used for BNP and NT-proBNP analysis respectively. Post-analysis, paired BNP and NT-proBNP results were categorised into four groups by patient age (<75y/>75y) and location (acute/non-acute). Acute category was comprised of in-hospital patients including ED and wards. Non-acute included patients from clinics and primary care. Concordance was evaluated using published decision thresholds (ESC, NICE) for ruling out HF in acute and non-acute settings, involving BNP (acute: <100, non-acute: <35 ng/L) and NT-proBNP (acute: <300, non-acute <125 ng/L). For NT-proBNP, a higher rule out threshold (450 ng/L) was applied to patients >75y irrespective of location.

Results

Of the 1048 samples, most were from patients <75y (acute: 42%, non-acute:31%), less than one third from patients >75y (acute: 12%, non-acute: 15%).Concordance was highest (91%) within the <75y non-acute group and least so (85%) within the <75y acute group. Discordant data was largely explained by patients with NT-proBNP above and BNP below respective rule out thresholds.

Conclusion

In conclusion our large study highlights discordance between BNP and NTproBNP, which is seldom reported, and was informative to the laboratory and clinicians during our implementation of NT-proBNP. Such discordance was largely explained by raised NT-proBNP which could only be explained in part for patients with reduced renal function.

Abstract 04

Verification of an Assay for Non-Esterified Fatty Acids and Evaluation of its Performance in Paediatric Hypoglycaemia Investigation

Gillick AJ1,2, Griffin M1, McConkey C1, Brennan C1, Carolan C1, Appleton M3 & Brady JJ1,2

1 Department of Clinical Biochemistry, Children’s Health Ireland at Temple St, Dublin 1,

2 School of Medicine, University College Dublin, Dublin, Ireland

3 Department of Biochemistry, Newcastle upon Tyne Hospitals NHS Foundation Trust, UK.

Background

Hypoglycaemia, a common paediatric emergency, has devastating neurological sequalae or mortality without recognition and treatment. Biochemical investigations during the hypoglycaemic episode are critical to establish the underlying cause. In hypoglycaemia, lipoprotein-lipase releases non-esterified fatty acids (NEFA) from triglycerides. Measuring NEFA in hypoglycaemic investigations helps to elucidate the cause and influence further testing and management. We aimed to verify the Wako NEFA-HR(2) assay on the SIEMENS Atellica® analyser and evaluate its clinical performance in hypoglycaemic investigations.

Methods

The method was verified according to CLSI guidelines. NEFA stability was investigated in 2 samples stored at room-temperature for 2, 4, 6, 24 and 48-h, 4 °C for 6, 24, 48 and 72-h, and −20 °C for 7, 14, 21 and 28-days, analysed in a single batch. The percentage difference from baseline NEFA for each condition was calculated and compared to total allowable error (14.85%). The effect of haemolysis and icterus was assessed by spiking 3 serum pools with red-cells achieving 10–1000 mg/dL haemoglobin or an icteric specimen to achieve an Atellica icteric index 3 (342–511 μmol/L bilirubin).

Results

Manufacturer precision claims (CV<1.5%) were confirmed at 1.38 mmol/L and 0.435 mmol/L. Linearity was confirmed to 2.6 mmol/L. LOQ was confirmed as 0.016 mmol/L. The method agreed with a manual WAKO assay (r2 0.98) and was within 2 standard deviations of the peer group mean in 3/5 EQA samples. Samples were stable for 6-h at room-temperature, 24-h at 4 °C and 28-days at −20 °C. There was no interference from bilirubin or heamoglobin up to 500 mg/dL. Between 500 and 1000 mg/dL haemoglobin decreases in NEFA of up to 51% were observed. NEFA concentrations supported the diagnosis in the hypoglycaemia investigations of 24/26 patients studied.

Conclusion

Our findings suggest the Wako NEFA-HR(2) method is robust and suitable for neonatal sample analysis with little interference from icterus and haemolysis. NEFA is clinically useful in paediatric hypoglycaemia investigations.

Abstract 05

Comparison of HbA1c measurement in remotely prepared capillary versus standard venous blood samples

Groenendijk W1, Griffin T2,3, Islam MN1, Blake L1, O’Shea PM1,3

1 Department of Clinical Biochemistry, Saolta University Health Care Group (SUHCG), Galway University Hospitals, Galway, Ireland.

2 Centre for Endocrinology, Diabetes and Metabolism, Saolta University Health Care Group (SUHCG), Galway University Hospitals, Galway, Ireland.

3 School of Medicine, National University of Ireland Galway (NUIG), Galway, Ireland.

Background

The COVID-19 pandemic has disrupted routine HbA1c testing. This has led to difficulties in monitoring glycaemic control and identifying people with suboptimal glycaemia. Delayed diagnosis of diabetes and suboptimal glycaemic control over extended periods can increase the risk of developing long-term complications of diabetes. The self-collection of capillary blood remotely (at home) for routine HbA1c testing can facilitate monitoring of glycaemic control whilst supporting virtual consultations. The aim of this study was to assess the clinical performance and user acceptance of capillary blood samples prepared remotely using the MiniCollect® capillary blood collection device as an alternative to standard venous blood collection for HbA1c analysis.

Methods

Adult men and women with any type of diabetes were recruited. Following informed written consent, eligible participants provided a venous blood sample at their routine clinic appointment and subsequently prepared a capillary blood sample remotely. Participants also completed a bespoke usability questionnaire.

Results

Of 84 participants recruited, 62 capillary samples returned to the laboratory, with 41 having a paired venous sample for HbA1c analysis. HbA1c results using both collection methods demonstrated good agreement; Passing–Bablok Regression analysis, y=0 + 1x; R=0.986, Bland–Altman Difference Plot providing a mean difference of 0.3 mmol/mol.

Conclusions

Over half of participants found the MiniCollect® device easy to use. The majority were in favour of the remote capillary blood collection service and would use it if routinely available. The remote self-collection of capillary blood for HbA1c is a convenient alternative for people with diabetes living and working in rural or urban settings ensuring optimal continuance of care.

Abstract 06

Evaluation of testosterone cut-offs for reflex testing for SHBG

Groenendijk W1, Green AD1, Corrigan H2, Lee GR1,2

1. Department of Clinical Biochemistry and Endocrinology, Mater Misericordiae University Hospital, Dublin, Ireland.

2. Department of Clinical Chemistry, Regional Hospital, Mullingar, Ireland.

Background

Testosterone measurement is an important investigation for the diagnosis and monitoring of several gonadal disorders. However, clinical interpretation requires consideration of potential alterations in protein binding, particularly with Sex Hormone Binding Globulin (SHBG). Free Testosterone (FT) (males) and Free Androgen Index (FAI) (females) are estimates of testosterone adjusted for SHBG level which can offer added clinical value to total testosterone (TT) alone. The aim of the present study was to assess the value of these testosterone cut-offs for reflexing SHBG.

Methods

In MRH, SHBG testing (Abbott Architect) is currently reflexed on any male TT (Beckman DXI) between 3 and 8 nmol/L (R: 6.1–27 nmol/L) and any female TT>1.5 nmol/L (LCMS, R: <1.9 nmol/L), with calculation of FT/FAI as appropriate. TT, SHBG and FT/FAI collected over a 2 year period were reviewed for males and females.

Results

In a total of 71 male patients, over a third (n=22) of TT results were clinically discordant with FT results. All patients (n=17) with a TT<6.1 (LRL) surprisingly had a low FT which was also lower for normal TT values up to 7.9 nmol/L. For the 79 female samples, almost one half (n=35/79) of results were discordant of which all were detected by the current reflex limit (TT>1.5 nmol/L). More than one third of patients (n=16/43) with TT>1.9 had a normal FAI. For TT results within the RR (<1.9), more than one half (n=20/36) had a raised FAI.

Conclusions

Our study overall on SHBG reflexing supports raising the lower (3–6 nmol/L) and potentially upper (7.9 nmol/L) TT reflex limit in males and lowering the cut-off for females. Ongoing review of a larger data is expected to better define new upper and lower TT thresholds for males and females respectively.

Abstract 07

Trimester-Specific-Reference Intervals for commonly requested biochemical and haematological parameters used in pregnancy care

Groenendijk W1, Dervan L2, Bogdanet D2,3, Finn O1, Islam MN1, Doheny H1, Dunne FP2,3, O’Shea PM1,2

1. Department of Clinical Biochemistry, Saolta University Health Care Group (SUHCG), Galway University Hospitals, Galway, Ireland.

2. School of Medicine, National University of Ireland Galway (NUIG), Galway, Ireland.

3. Centre for Endocrinology, Diabetes and Metabolism, Saolta University Health Care Group (SUHCG), Galway University Hospitals, Galway, Ireland.

Background

When reporting laboratory parameters on pregnant women, the reference intervals (RIs) provided should be specific to pregnancy and not the non-pregnant adult population. This is because results have different significance depending on the stage (trimester) of pregnancy. The aim of the current study was to establish trimester-specific-RIs in healthy pregnant women.

Methods

A prospective cross-sectional study of apparently healthy pregnant volunteers was conducted. At enrolment, inclusion criteria were: signed informed consent, age≥18 years, White European, body mass index<25 kg/m2, blood pressure<140/90 mmHg, non-smoker, no previous or new diagnosis of pathology. Trimester was defined as, T1: up to 13 weeks + 6 days, T2: 14–27 weeks + 6 days, T3: >28–41 weeks + 6 days).

Results

Baseline demographics, anthropometric and laboratory measurements were recorded. In total, 31 biochemical and 10 haematological ISO15189:2012 accredited tests were analysed using the Roche Cobas® and Siemens Advia® 2120i platforms respectively following standard operating procedures. RIs were defined according to the International Federation of Clinical Chemistry (IFCC) recommended method. Ostensibly healthy pregnant women (n=1232) were recruited, 953 (77.4%) failed the inclusion criteria.

Abstract 08

Therapeutic potential of parasite-derived immunomodulatory peptides in the treatment of sepsis

Hamon SM1, Griffin MD2, Dalton JP1 & Lalor R1

1. Molecular Parasitology Laboratory, Centre of One Health, Ryan Institute, National University of Ireland Galway (NUIG)

2. Regenerative Medicine Institute (REMEDI), NUIG

Background

In Ireland, >16,000 people develop sepsis annually, with over 17% of them dying in hospital. Sepsis deaths result from excessive inflammatory responses to infectious microbes. The failure of the normal protective functions of the immune system allows for intense inflammation that causes damage to organs which can be fatal. Helminth (worm) parasites infect almost two billion people worldwide. They survive for decades in the body by releasing molecules that modify how the immune system functions. We discovered that peptides released by the helminth, Fasciola hepatica (liver fluke), termed Helminth Defense Molecules (HDMs), dampen their host’s inflammatory response, in particular the responses of immune cells. The aim is to determine the effectiveness and mechanism of peptides derived from Fasciola hepatica, namely FhHDM C2, in the treatment of sepsis. FhHDM C2 interferes with Toll-like Receptor 4 (TLR4) preventing a cascade of signalling that impedes the production of pro-inflammatory cytokines such as TNFα.

Method

Human peripheral blood mononuclear cells (PBMCs) (n=4) (License: 034-001-21) were isolated by Lymphoprep density gradient medium and added to tissue culture plates. Cells were allowed rest for 1 h at 37 °C and 5% CO2, next FhHDM C2 (5 μM or 10 μM) was added for 1 h at 37 °C and 5% CO2. Lipopolysaccharide (LPS) (100 ng/mL) or peptidoglycan (PGN) (10 μg/mL) was added for 18 h and incubated at 37 °C and 5% CO2. Supernatants were analysed by ELISA for TNFα, IL10 and IL1β.

Results

FhHDM C2 (10 μM) significantly reduced pro-inflammatory cytokines TNFα, IL10 and IL1β (p=<0.05) after stimulation with LPS. However, PGN, a TLR2 ligand, significantly increased pro-inflammatory cytokines TNFα, IL10 and IL1β (p=<0.05).

Conclusions

FhHDM C2 demonstrates promising anti-TLR4 effects in human immune cells and potential as a sepsis treatment. However, the cytokine results observed post incubation with PGN warrant further investigation to elucidate its overall significance in sepsis.

Abstract 09

Measurement of glycated CD59 (gCD59) in plasma: Development and validation of a method

Islam MN1, Bogdanet D2, Doheny H1, Groenendijk W1, Dunne F2,3 & O’Shea PM1,3

1. Department of Clinical Biochemistry, Saolta University Health Care Group (SUHCG), Galway University Hospitals, Galway, Ireland.

2. Centre for Endocrinology, Diabetes and Metabolism, Saolta University Health Care Group (SUHCG), Galway University Hospitals, Galway, Ireland.

3. School of Medicine, National University of Ireland Galway (NUIG), Galway, Ireland.

Background

Diagnosis of gestational diabetes (GDM) is carried out by the Oral Glucose Tolerance Test (OGTT) at 24–28 weeks’ gestation according to the International Association of Diabetes in Pregnancy Study Group criteria. However, OGTT has many limitations some of that include poor reproducibility, time-consuming and inconvenience. Recent studies highlighted glycated CD59 (gCD59) as a potential alternative to OGTT based GDM diagnosis (1). However, gCD59 measurement is relatively complex and the assay is not commercially available yet. This study aimed to optimise and validate a simplified method for measuring plasma gCD59.

Method

In this study we evaluated two methods for separation of gCD59 from total CD59 in plasma. First method utilises BRIC-229 antibody (binds only to non-glycated CD59) and second method utilises Sodium borohydride (NaBH4; binds only to gCD59). Based on performance of those two methods, NaBH4-based method was applied to measure plasma gCD59 using a commercial CD59 enzymelinked-immunosorbent-assay. This study also aimed to compare our method with previously published method. Plasma gCD59 levels were measured in pregnant cohorts at trimester-1 (T1; <14 weeks of gestation) and trimester-2 (T2; 24–28 weeks of gestation) from 50 women with GDM and 100 nondiabetic pregnant women controls who were matched for age, BMI, ethnicity and gestational age.

Results

Precision/bias study confirmed overall assay imprecision of 5.6%, 2.5% and 2.1% and overall bias of 0.67%, 0.15% and 1.28% with mean CD59 concentrations of 31.5 pg/mL, 125.2 pg/mL and 493.6 pg/mL for actual CD59 concentrations of 31.25 pg/mL, 125 pg/mL and 500 pg/mL respectively. Plasma gCD59 concentrations did not vary between controls and GDM cohorts (T1: median 3.71 ng/mL vs. 3.71 ng/mL; p=0.60 and T2: median 4.24 ng/mL vs. 4.20 ng/mL, p=0.74). Receiver Operator Curve analyses confirmed that both the newly developed assay and the published method were comparable.

Conclusion

To conclude, this study developed a simplified precise method for plasma gCD59 measurement.

Reference

  • 1. Ghosh P et al. (2017) Plasma Glycated CD59, a Novel Biomarker for Detection of Pregnancy-Induced Glucose Intolerance. Diabetes Care. 40, 981–984.

Abstract 10

Male Pituitary-Gonadal Axis Function in Obstructive Sleep Apnoea Syndrome: The Effect of Continuous Positive Airway Pressure

Leigh R1, Hamon SM2, McWeeney M3, O′ Shea PM4 & O′ Loughlin A1

1 Bon Secours Hospital, Galway.

2 Ryan Institute, NUIG.

3 Galway Clinic, Doughiska.

4 Clinical Biochemistry Department, Galway University Hospital.

Background

Obstructive sleep apnoea syndrome (OSAS) is common; is proportionately affecting the overweight and obese. Continuous positive airway pressure (CPAP) is the first-line treatment for moderate to severe OSAS. Clinical equipoise exists as to whether CPAP treatment directly affects pituitary-gonadal hormone synthesis. This study aimed to determine the effect of CPAP treatment on gonadotropins, prolactin, sex-hormone binding-globulin (SHBG), total testosterone (TT) and calculated free testosterone (cFT) in male OSAS patients.

Methods

Ethical approval was sought from the Bon Secours Health System Clinical Ethics Committee (CEC) and granted on the 08/03/2019. All participants in the research gave informed consent, obtained in accordance with guidelines provided by the CEC, Trinity College Dublin and the RCSI. Participants provided venous blood samples before and twice after (first night CPAP, n=25 and 3 months CPAP, n=13) commencing CPAP treatment. At each time-point, concentrations of TT, SHBG, prolactin and gonadotropins were measured. In total, 53 males with a diagnosis of OSAS confirmed by polysomnography were prospectively enrolled to this study.

Results

Hypogonadism in the cohort was low (n=2). Hyperprolactinaemia was prevalent (n=25). TT and cFT were significantly negatively correlated with obesity. cFT was correlated with OSAS severity, but not TT. Significant reductions were observed in TT (pre 16.6 nmol/L, post 13.5 nmol/L, p=0.003), cFT (pre 332 pmol/L, post 250 pmol/L, p=0.001) and prolactin (pre 360 mIU/L, post 225 mIU/L, p=0.006) after 3-months of CPAP (n=13). No significant change was observed in other pituitary hormones or SHBG.

Conclusion

The prevalence of hypogonadism is low in this cohort. CPAP treatment reduced testosterone and prolactin in eugonadal males with OSAS. The benefits of CPAP treatment for OSAS may be independent to change in serum testosterone levels. Hypogonadal OSAS patients should be managed via strategies other than CPAP alone.

Abstract 11

Investigation of the relationship between serum amyloid-beta 42 concentration and Alzheimer disease in south-west Ireland

Purcell E, Molloy WJ & Costelloe SJ

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

Background

Alzheimer disease (AD) is the leading cause of dementia. Current diagnostics for AD can be costly or invasive. Amyloid-Beta 42 (Aβ42) has proven clinical utility when measured in cerebrospinal fluid for the diagnosis of AD. Measurement of Aβ42 in serum would be less expensive, less invasive, and more accessible. This study sought to evaluate the use of Aβ42 in serum as a diagnostic biomarker for AD.

Methods

Two cohorts were compared. One cohort consisted of 30 patients (male 12, female 18, median age 79.9) with known AD attending a memory clinic in St. Finbarr’s Hospital Cork. The second cohort consisted of 41 healthy individuals (male 10, female 30, median age 33.2) employed in Cork University Hospital. Serum samples were analysed by the Assay Genie manufactured enzyme linked immunosorbent assay (ELISA) for Aβ42 on the Dynex DS2. Quick mild cognitive impairment (Qmci) score was used to assess cognition. Chi-squared analysis examined the relationship between concentrations of Aβ42 and cognitive impairment – subgroup analysis compared Aβ42 with age.

Results

The median (IQR) concentration of Aβ42 was 19.66 pg/mL in the control group, 83.49 pg/mL in the study group. Chi-squared analysis demonstrated a statistically significant relationship between Aβ42 and AD (p=0.008). No significant relationship was found between Aβ42 and age, Chi-squared analysis (p=0.364).

Conclusions

Aβ42 was significantly higher in the AD group than in the control group. Several factors need to be addressed before this test can be considered for diagnostic purposes, including full validation of a method suitable for clinical use. The authors suggest the analysis be rerun with an age-matched control group. The results are promising and a basis for future research into serum Aβ42 as a valuable diagnostic tool in AD.

Abstract 12

Quantifying the effect of ethylenediamine tetra acetic acid (EDTA) contamination on certain biochemistry analytes

Ramzan Matthews S1, Costelloe, SJ1 & Rico Rios, N1

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

Background

The order-of-draw (OOD) requires serum samples to be taken before ethylenediaminetetraacetic acid (EDTA) additive tubes. When OOD is not followed, it can lead to potential EDTA contamination. This causes typical biochemistry patterns were raised potassium and decreased calcium, magnesium, alkaline phosphatase (ALP) and iron concentrations are seen. This pattern is evident in gross EDTA contamination but subtle degrees of contamination may be unnoticed. Hence, this study will look at which biochemical analytes are first affected with increasing concentrations of EDTA. This study assesses biochemical analytes known to be affected by EDTA contamination to see at what concentration of EDTA their concentrations are affected.

Methods

Ten EDTA contamination solutions with concentrations ranging from 0.1 to 1 mmol/L were prepared using serumEDTAfree and spiked with different volumes of a 20 mmol/L EDTA stock solution. SerumEDTAfree and EDTA contamination solutions were analysed for biochemistry parameters affected by EDTA contamination (Calcium, ALP, Magnesium, Iron). Potassium was not evaluated as EDTA used was pure. Parameter concentrations were compared in serumEDTAfree vs. each EDTA contamination solution and relative differences (%) were calculated.

Results

Calcium and iron showed a significant gradient negative (%) difference that correlated with the increasing EDTA concentration: Calcium from −15.88% (EDTA=0.1 mmol/L) to −51.50% (EDTA=1 mmol/L); Iron from −75.86% (0.1 mmol/L) to −89.08% (1 mmol/L). Magnesium showed less significant gradient, from −9.28% (0.1 mmol/L) to −29.90% (1 mmol/L). ALP didn’t show a difference with gradient concentrations.

Conclusions

This study showed a clear effect of increasing EDTA concentrations on iron and calcium which both decreased in a gradient fashion with increasing EDTA concentration. The unexpected pattern of results for magnesium and ALP may depend on the initial concentration of the analytes in serum or may depends on a delay in the EDTA effect seen with magnesium and ALP due to magnesium acting as a co-factor for ALP activity.

Abstract 13

Detection of non-adherence to the correct phlebotomy order-of-draw using routine biochemical data

Ramzan Matthews, S1, Bransfield, A1, Costelloe, SJ1 & Rico Rios, N1

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

Background

Contamination of serum or plasma specimens with ethylenediaminetetraacetic acid (EDTA) is a well-known problem in laboratory medicine. Although gross EDTA contamination is readily detected by the skilled laboratorian, subtle contamination may go unnoticed unless laboratories have access to an EDTA assay. Poor adherence to the correct order-of-draw (OOD) is more likely to lead to EDTA contamination. This study aimed to compare the incidence of biochemical abnormalities associated with EDTA contamination and assess the frequency of it by specific hospital locations where an EDTA was taken at the same phlebotomy episode (Serumpaired) compared with serum where no EDTA was collected at the same time (Serumalone). This may prove a useful tool for laboratories to identify requesting locations who are failing to adhere to the correct OOD.

Methods

Data were gathered for a 9-month period from laboratory databases at Cork University Hospital and analysed in Excel and RStudio. The mean concentrations of potassium, calcium, magnesium, alkaline phosphatase (ALP) and iron in the Serumpaired samples were compared with those in the Serumalone samples using histograms.

Results

Of 1,041,396 samples analysed, 60,116 (6%) were Serumalone samples while 981,280 (94%) were Serumpaired samples. Mean concentrations and standard deviation (SD) from serumpaired and serumalone specimens were compared: Calcium (serumalone=2.4 ± 0.18 μmol/L), serumpaired=2.3 ± 0.16 μmol/L, (P=0.934); Iron (serumalone=18.1 ± 10.51 μmol/L, serumpaired=16.1 ± 9.47 μmol/L (P=0.556); Magnesium serumalone=0.9 ± 0.24 μmol/L, serumpaired=0.8 ± 0.16 μmol/L, (P=0.556); ALP (serumalone=103 ± 84.22 μmol/L, serumpaired=98.6 ± 86.72 μmol/L, (P=0.219); Potassium (serumalone=4.5 ± 0.69 μmol/L, Serumpaired=4.4 ± 0.57 μmol/L, (P= 0.422).

Conclusions

This tool is not a conclusive method of identifying potential EDTA contamination by specific locations. Although results demonstrate that there may be some subtle EDTA contamination, these are not statistically significant due to the small number of serumalone samples when compared to serumpaired samples.

Abstract 14

Non-adherence to the correct order-of-draw and the effect on biochemical parameters; an experimental study

Matthews Ramzan S1, Rico Rios N1, Andrade N1, O’Donovan K2 & Costelloe SJ1

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

2 Emergency Department, Cork University Hospital, Wilton, Co. Cork, Republic of Ireland

Background

The order-of-draw (OOD) is a universally agreed convention by which specimens are drawn in a specific order to avoid cross-contamination between tubes containing different additives. Serum samples grossly contaminated with additives such as ethylenediaminetetraacetic acid (EDTA) and fluoride oxalate can yield characteristic patterns identifiable in the post-analytical phase. However, contamination may be subtle, leading to spurious results and adverse patient care. This study investigates the effect of the incorrect OOD on the concentration of biochemical parameters known to be susceptible to EDTA contamination.

Methods

A previous survey amongst Emergency Department staff identified the five most common errors in adherence to the OOD, these were assigned in five groups (G1–G5). Bloods were collected from five volunteers per group. In each case, an SST was taken first as a control and an SST was taken as a final tube. Both SSTs samples were analysed for calcium, magnesium, ALP, iron and potassium. Box and whisker plots and chi-square statistical analysis were used to compare each parameter across the five groups.

Results

Median concentrations and interquartile range (IQR) on control vs. final SST resulted as follows: Calcium 2.45 (0.03) vs. 2.43 (0.08) mmol/L (G4); Magnesium 0.89 (0.05) vs. 0.86 (0.11) mmol/L (G2); Potassium 4.40 (0.20) vs. 4.60 (0.30) mmol/L (G2), 4.30 (0.35) vs. 4.40 (0.38) mmol/L (G3), 4.30 (0.10) vs. 4.40 (0.33) mmol/L (G5). No statistical significant difference (p<0.05) was found on each sample when comparing concentrations between final and control SSTs. However, there was an expected average difference of −1% (calcium, G1), −2% (iron, G2), +4% (potassium, G2, G5); and a paradoxical difference of +1.5% and +2.4% (magnesium, G4, G5), +1.7% (iron, G4), −3.2% (potassium, G4).

Conclusions

Expected EDTA contamination effect was found for magnesium, calcium and potassium, but paradoxical results were also obtained. Potential reasons may include: the same phlebotomist using the same technique performed all collections and additive tubes were not filled fully. Non-adherence to the OOD does not always result in appreciable levels of EDTA contamination that affect biochemical parameters.

Abstract 15

Assessment of sample type and preanalytical factors on homocysteine concentration in metabolic patient samples

Talbot C, Sheerin A and Lee GR.

Department of Clinical Chemistry and Diagnostic Endocrinology, Mater Misericordiae University Hospital

Background

Total homocysteine (tHcy) levels in patients with homocystinuria (HCU) provide a biochemical target for dietary therapy. Analysis of tHcy levels is complicated by pre-analytical factors that limit the provision of tHcy analysis only to samples that are collected on site and centrifuged in a timely manner. The Sarstedt Monovette Hcy-C tube (containing sodium citrate and buffer) is specifically designed for homocysteine analysis with increased stability of tHcy in samples up to a maximum of 48 h at 4 °C. We investigated the effect of the Sarstedt Monovette Hcy-C tube type and preanalytical factors on homocysteine concentration in patients with HCU and the possibility of off-site tHcy sample collection.

Methods

Samples were collected from 26 study participants (21 HCU patients and 5 volunteers). Each participant provided a lithium heparin sample on ice and two Hcy-C sample tubes. tHcy was analysed by immunoassay on the day of collection in paired lithium heparin and Hcy-C samples. A second Hcy-C tube was left un-processed at 4 °C or −20 °C for between 24 and 48 h before tHCys analysis, to examine effect of storage and freeze-thawing on tHcy concentration.

Results

tHcy concentrations were only marginally lower (mean=4.3%) lower in samples collected in Hcy-C compared to lithium heparin tubes. tHcy concentrations did not change (P>0.05) when samples collected in Hcy-C tubes were left unprocessed for 24hrs at 4 °C and decreased marginally (P>0.05, 3.4%) 3.4% following a freeze-thaw cycle.

Conclusions

Our study suggest that the Hcy-C sample tubes could permit longer periods of storage time postvenepuncuture, given the minimal effect on tHcy concentrations. This represents a more feasible and flexible approach for patients to collect samples without attending hospital phlebotomy.

Abstract 16

Pre-analytical centrifugation delay: a retrospective study in a regional biochemistry laboratory

Vinod S1, Coleman-Vaughan C1, Rico Rios N2, Costelloe SJ2

1. Department of Biological Sciences, Munster Technological University Cork, Ireland.

2. Department of Clinical Biochemistry, Cork University Hospital, Republic of Ireland.

Background

One of the most important factors affecting the stability of an analyte is delayed centrifugation. Most analytes are unaffected by this and will remain stable in whole blood for an extended period. However, when in prolonged contact with cells, analytes such as potassium and sodium will result in artifactual alterations in their concentrations. These values can often still be interpreted as within the normal range but are not representative of the patient’s actual biological status. This study investigated what proportion of samples sent for routine biochemical testing were centrifuged within their pre-centrifugation stability limit. Furthermore, the effect that delayed centrifugation had on these specimens and whether spurious results were reported as a consequence was assessed.

Methods

Renal profile (potassium, creatinine, chloride, sodium, urea), liver profile (total protein, albumin, alanine aminotransferase, alkaline phosphatase, and total bilirubin), bone profile (calcium, phosphate) and miscellaneous requests (c-reactive protein (CRP), lactate, lactate dehydrogenase, oestradiol and magnesium) received into the clinical biochemistry laboratory of Cork University Hospital (CUH) over one month were examined to determine if they were processed within their pre-centrifugation stability limits.

Results

Within the 53,997 specimens received, all non-GP requested tests had an error rate of <2% per analyte, except for lactate and oestradiol, which were 3.8 and 8.1%, respectively. From all tests requested by GP’s, 17.55% of chloride, 15.45% of potassium, 15.74% of phosphate and 18.82% of CRP were processed beyond their pre-centrifugation stability limits. A total of 4,881 patients had results returned where samples were processed outside of their pre-centrifugation stability limit.

Conclusions

Adding a new rule to the lab information system that can correctly identify samples that were not centrifuged within the pre-centrifugation stability limit of the analyte of interest could decrease the likelihood of samples that have been received outside these limits being further processed.

Abstract 17

Psychiatric Presentations of Inborn Metabolic Disorders in Older Children

Fitzsimons PE1, Durkan Á1, Shyne V1, G Urbano1, Lynch B2, Treacy EP3,5, Monavari AA4,5, Knerr I4,5, & Brady JJ1,5

1. Departments of Paediatric Laboratory Medicine, Children’s Health Ireland at Temple Street.

2. Paediatric Neurology, Children’s Health Ireland at Temple Street.

3. Adult Metabolic Unit Mater Misericordiae University Hospital

4. National Centre for Inherited Metabolic Disorders at Children’s Health Ireland

5. School of Medicine, University College Dublin.

Background

Inherited metabolic disorders can manifest as behavioural changes, acute intermittent attacks of psychosis or hallucinations in older children, adolescents or adults. In some cases because of intermittent presentation they may be challenging to diagnose. We present a case series where acute behavioural changes and/or psychosis were presenting features in three older children that led to appropriate biochemical investigations and ultimate diagnosis.

Case 1: An 8 year-old girl with a background history of vomiting, recurrent UTI, poor coordination presented with acute onset of confusion, erratic behaviour and agitation. Initial biochemical findings showed transaminitis, negative toxicology screen. Plasma ammonia was 324 umol/L (ref<65).

Case 2: A 14 year-old boy with a background history of noctambulism presented with acute onset of confusion and erratic behaviour. Toxicology screen was negative. Initial biochemical findings showed mild transaminitis. CRP was normal. Plasma ammonia was 200 umol/L. Plasma amino acids in both Case 1 & 2 showed increased glutamine, low/normal ornithine and arginine. Urine organic acids showed an elevated orotate. Findings were consistent with a urea cycle defect, Ornithine Transcarbamylase (OTC) deficiency.

Case 3: A 10 1/2 year-old girl with a background history of learning disability and focal seizures presented acutely with psychiatric symptoms, including agitation, hallucinations and episodes of screaming and aggression. Plasma amino acids showed detectable free homocystine with low methionine, total homocysteine was 330 umol/L (ref 3–10). CSF methionine was not detected. Amino acid results were suggestive of 5,10-methylene tetrahydrofolate reductase (MTHFR) deficiency. Low serum folate, normal MCV, normal methylmalonic acid and normal propionylcarnitine made an intracellular cobalamin-related remethylation disorder less likely. MTHFR deficiency was genetically confirmed.

Conclusions

These cases highlight that baseline biochemical and metabolic investigations should be performed alongside toxicology in the setting of an acute psychiatric/neurological presentation in any age group to allow for appropriate recognition and timely intervention.

Abstract 18

It’s in the Bag! Pre-analytical cause of spurious electrolytes

Kavanagh-Wright L & Lee GR

Dept. of Clinical Chemistry and Diagnostic Endocrinology, Mater Misericordiae University Hospital

Background

Therapeutic venesection is prescribed to treat iron overload in cases of primary and secondary haemochromatosis. Approximately one pint of blood is removed per appointment, until an acceptable Ferritin concentration is reached (20–100 ug/L). The aim is to then keep Ferritin at this maintenance level to prevent the signs and symptoms of iron overload. In most cases the blood bags are discarded after venesection. Some bags are fitted with a sampling portal to facilitate Ferritin levels to be measured without having to take further blood samples. We present two cases of spurious biochemistry results on analysis of “Li-Hep plasma” for renal and bone profiles.

Case 1: A 58 year old male, attended our Oncology Haematology Unit. He required venesection for iron overload, secondary to multiple red cell transfusions. Analysis showed a raised Sodium (149 mmol/L) without concurrently raised Chloride, raised Phosphate (2.18 mmol/L) and high anion gap (35).

Case 2: A 69 year old male, attended our Liver Centre, for therapeutic venesection for HFE Hereditary Haemochromatosis. Analysis showed a markedly raised Sodium (167 mmol/L) without concurrent raised Chloride, low Calcium (1.84 mmol/L), markedly raised Phosphate (3.57 mmol/L) and markedly high anion gap (58).

Conclusions

The dispropowrtionately elevated Sodium and high Anion Gap in both cases caused suspicion of sample contamination. When the Duty Biochemist contacted the clinical teams, it was ascertained that samples were not taken from peripheral vein but rather from the venesection bag via the sampling portal. Preservatives in such venesection bags include Monosodium Phosphate, Sodium Citrate and Dextrose. These preservatives can subtly or markedly contribute to spuriously raised sodium, glucose and phosphate, while the citrate anion may cause low calcium by its chelating action and contribute to a high anion gap. Venesection bags are likely an under-reported cause of spurious biochemistry results which should be highlighted.

Abstract 19

Pseudohypoaldosteronism Secondary to Renal Abnormalities in an Infant

O’Dwyer J1, Abdelrahman S2, Molloy M1 & Brady JJ1,3

1. Department of Paediatric Laboratory Medicine, Children’s Health Ireland at Crumlin, Dublin 12.

2. Department of General Paediatrics, Children’s Health Ireland at Crumlin, Dublin 12.

3. Department of Paediatric Laboratory Medicine, Children’s Health Ireland at Temple Street, Dublin 1.

Case:

A four-month old female presented to ED with intermittent vomiting, low intake for one week and lethargy for the previous few days. The patient had lost weight since birth and was now on the 2nd centile. She appeared pale, afebrile, and blood pressure of 147/80 mmHg. Initial blood gas analysis showed a non-anion gap metabolic acidosis with severe hyponatraemia and hyperkalaemia; (pH=7.356, pCO2=3.95 kPa, BE=−7.9 mmol/L, HCO3=17.9 mmol/L, Na=115 mmol/L, K=7.6 mmol/L, Cl=85 mmol/L). Urine analysis showed Na=26 mmol/L, K=3.1 mmol/L, Cl=25 mmol/L. Possible causes included congenital adrenal hyperplasia, adrenal insufficiency and hypoaldosteronism. Normal skin pigmentation, normal blood glucose and no virilisation made CAH due to 21-hydroxylase deficiency unlikely; confirmed by a 17- hydroxyprogesterone concentration of 4 nmol/L (RR<13 nmol/L). Cortisol (1156 nmol/L [RR 115–450 am]) and ACTH (16.1 ng/L [RR 7–63 ng/L]), were normal out ruling ACTH insufficiency. The raised blood pressure was not compatible with a diagnosis of hypoaldosteronism. Urine cultures were positive for Escherichia coli which suggested the possibility of pseudohypoaldosteronism (PHA) secondary to urinary tract infection (UTI). Urinary steroid profile ruled out salt wasting forms of CAH and aldosterone synthase deficiency; it also noted an increased concentration of free cholesterol; a finding seen in PHA secondary to UTI. Fluid resuscitation and calcium gluconate were given and electrolytes normalised. No hydrocortisone was given. Coamoxiclav and gentamicin were prescribed to treat the underlying infection.

Conclusions

Inflammation due to the UTI is suspected to cause aldosterone resistance. In many cases, UTI is an intercurrent illness due to a urinary tract anomaly. Renal imaging in this child showed focal pyelonephritis of the right kidney and urine reflux in both kidneys. This case illustrates a relatively common scenario in the paediatric setting which must be considered in a child presenting with hyponatraemia and hyperkalaemia.

Abstract 20

Negative interference in creatinine measurement following paracetamol overdose

Smith R1, Matthews Ramzan S1, Owens S1, Dillon A2, Costelloe SJ1 & Joyce CM1

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

2. Department of Emergency Medicine, Cork University Hospital, Ireland.

Background

A 54y woman presented to the Emergency Medicine department at Cork University Hospital (CUH) with suspected paracetamol overdose. Biochemical analysis on admission included electrolytes, renal and liver function tests, paracetamol, salicylate, amylase, lactate and blood gasses. A paracetamol overdose was confirmed and N-acetylcysteine (NAC) was administered 3 h post admission. The patient’s creatinine concentration was critically low but as there was a high clinical suspicion for renal impairment, negative interference in the creatinine assay was considered. Investigation of a sample with suspiciously low creatinine concentration across different assays and analytical platforms.

Method

The Beckman Coulter enzymatic creatinine assay was performed on the AU58300 analyser at CUH. Bloods on admission and following treatment with NAC were sent to the laboratory for analysis. Paracetamol metabolites N-acetyl-p-benzoquinone imine or NAC can interfere with peroxidase formation in the Trinder reaction and generate a falsely low creatinine result. Creatinine results for this patient were compared using the Beckman Coulter, Roche, and Abbott enzymatic assays and the Abbott Jaffe assay.

Results

On admission, the patient’s creatinine was very low (14 μmol/L) with slightly elevated urea (11.1 mmol/L) levels and elevated paracetamol (780 mg/L). NAC was given 3 h post admission and serum creatinine when measured 5 h post admission was undetectable (<4 μmol/L). Reanalysis of this sample on the Abbott and Roche enzymatic assays showed they were less prone to negative interference with results of 91.9 and 125 μmol/L, respectively. The Jaffe method was least susceptible to negative interference with a creatinine result of 152 μmol/L.

Conclusions

Our study demonstrates that the Beckman Coulter enzymatic creatinine assay is most sensitive to interference following paracetamol overdose. We recommend cautious interpretation of enzymatic creatinine results in this context and advise reanalysis using the Jaffe reaction or point of care devices to monitor renal function.


*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.


Published Online: 2023-03-06
Published in Print: 2023-05-25

© 2023 Walter de Gruyter GmbH, Berlin/Boston

Downloaded on 28.3.2024 from https://www.degruyter.com/document/doi/10.1515/cclm-2023-0192/html
Scroll to top button