WCS10 The Use Of Laboratory Test In Clinical Medicine Flashcards
Laboratory Testing Cycle
Decision to perform test —>
Pre-analytical phase: —> Test request —> Identification + essential clinical information entered —> Order + specimen transferred to lab —> Specimen accessioned + processed
Analytical phase:
—> Specimen analysed
Post-analytical phase: —> Result generated, validated + authorised by lab —> Result reported to clinician —> Data interpreted —> Clinical response to result
Questions to ask
- Why request this test?
- What are consequences of not performing the test?
- How good is the test discriminating between health and disease?
- What to look for in result?
- How are results interpreted?
- How will test results influence patient management and outcome?
- Will investigation ultimately benefit the patient?
Purpose of laboratory test
- Confirm diagnosis (e.g. T4, TSH in suspected hyperthyroidism)
- Aid DDx (e.g. Distinguish between different forms of jaundice)
- Refine a diagnosis (e.g. Use of ACTH to differentiate causes of Cushing syndrome)
- Monitor disease progress (e.g. Plasma glucose + K to follow treatment with DKA, tumour markers after surgery)
- Assess severity (e.g. Serum creatinine + Urea in renal disease)
- Detect complications of disease + SE of treatment (e.g. ALT/AST in hepatotoxic drugs)
- Monitor therapy (e.g. drug blood level: anti-convulsants, immunosuppressants)
- Stratify risk of developing a disease for preventive therapy
- Population screening
Pharmacogenetic + Genomic testing
Prediction of Adverse drug effects / Therapeutic efficacy (personalised / precision medicine)
—> Maximise therapeutic benefits + Minimise unwanted drug effects
- HLA-B*1502: Carbamazepine induced SJS/TEN
- HLA-B*5801: Allopurinol-induced severe cutaneous adverse reactions
- Molecular markers guiding ***targeted therapy for cancers
- Dihydropyrimidine dehydrogenase deficiency (DPD) associated with enhanced toxicity of **5-FU and **Fluoropyrimidine capecitabine in affected subjects who have reduced drug clearance
Examples of test used in Case-finding programme
Neonates:
- Congenital hypothyroidism: Cord blood TSH +/- FT4
- G6PD deficiency: G6PD
Adolescents and young adults:
- Substance abuse: Urine drug screen
Pregnancy:
- Diabetes: Plasma glucose, OGTT
- Open neural tube defect in fetus: Maternal serum α-fetoprotein
Industry:
- Lead exposure: Whole blood lead levels
- Organophosphate pesticides exposure: Serum pseudocholinesterase activity
Elderly:
- Malnutrition: Serum Vit D, prealbumin, retinol binding protein
- Thyroid dysfunction: Serum TSH +/- FT4
Population screening
Key considerations:
- Condition is common / life-threatening
- Tests readily applicable and acceptable to the population
- Sensitive and Specific
- Facilities are available for subsequent follow-up and confirmation
- Economic impact has been assessed and implication accepted
Screening for rare diseases
High sensitivity, Low false-positive required
—> 95% reference interval not applicable since rate of false positive would be too high
—> Diagnostic cut-off value need to be adjusted to reduce false-positive rate
***Performance of diagnostic tests
Predictive values: depend on ***Prevalence
PPV: ↑ with Prevalence (True positive / All positive)
NPV: ↓ with Prevalence (True negative / All negative)
Performance: depends on ***Cut-off value chosen —> Sensitivity / Specificity vary with cut-off value chosen
Receiver-Operating Characteristic (ROC) curve: display “True positive” vs “False positive” across a range of cut-off values
—> Sensitivity vs (1-Specificity)
—> enable **selection of optimal cut-off for clinical use
—> **largest AUC: good
Performance can be improved by:
- Increase sensitivity
Laboratory result interpretation
In light of:
- Clinical context
- Knowledge of various factors that may affect the results
Establishing diagnosis based on laboratory results
4 Principal approaches:
- Hypothesis deduction (most common)
- list of DDx by history / physical exam —> selection of laboratory tests to confirm - Pattern recognition
- Medical algorithms
- Rifles vs Shotgun approach
Interpretation of clinical lab tests
- Normal / Abnormal
—> Reference intervals needed
—> Mean +/- 2SD (data with Gaussian distribution)
—> 2.5-97.5 centiles (i.e. 5% of normal subjects may have results outside the interval)
—> ***deviation from RI may not mean presence of disease
—> should always be interpreted in proper clinical context - Result fit with previous assessment of patient?
—> can explain the discrepancy if any? - Pre-test (i.e. Prevalence) + Post-test probability of presence / absence of disease
- Any significant change occurred in any of the results?
- Any of the results alter my diagnosis of patient’s illness / influence way of management?
- If cannot explain result, what do I propose to do about it?
- Serial results
- observed change due to imprecision? Or change in patient’s condition?
- under identical conditions, differ by **>2.77 times of total SD —> >95% it is a statistically significant change
- whether clinically significant: depends on **analyte + ***clinical context
- if difference > 1.96 x √2 x √(CVa^2 + CVb^2) —> <5% chance due to random variation
- CVa: Analytical variation
- CVb: Biological variation
***Major causes of interference in lab test
- Lipids, haemoglobin, other serum constituents (e.g. bilirubin)
- Anti-reagent Ab (e.g. heterophilic Ab, rheumatoid factor)
- Anti-analyte Ab (e.g. anti-thyroglobulin Ab)
- Other interferents (e.g. biotin)
- Macro-complexes (e.g. macro-prolactin, macro-TSH, macro-enzymes)
- Paraproteins
Clues for suspecting assay interferences
- Inconsistent with other lab test results
- Unusual in more than 1 assay (esp. in immunoassays)
- Significantly and inexplicably changed in comparison with previous results (delta-check, transversal and longitudinal data assessment, plausibility checks)
- Grossly and persistently “abnormal” results in an apparently healthy subject
- Clinically unexpected in light of clinical context
- Inconsistent with other clinical correlates
***Pre-analytical factors leading to Artefactual results
- Prolonged venous stasis
—> ↑ total Ca, Albumin, Lipid - Stress
—> ↑ Prolactin, ACTH, Cortisol - Smoking
—> ↑ Ammonia - ***Excess Heparin
—> ↓ ionised Ca, electrolytes - ***EDTA contamination
—> ↓ Ca, ALP, Mg, ↑ K - Alcohol swab
—> ↑ Ethanol (isopropanol) - Li-heparin tubes
—> ↑ Li - Collection from a Drip-arm, inadequate priming from venous / arterial blocks
—> Erroneous Al, Zn, Mn, Se —> **Acid-washed tubes needed
—> Ionised Ca —> **Lyophilised Heparin Tubes at Titrated dose needed
Erythrocytes in vitro changes
Take up:
- Na
- Cl
- ***Glucose (越黎越少) —> Glycolytic enzyme inhibitor + Prompt separation of plasma from cells
Release:
- K
- CO2
- HCO3
