Laboratory Medicine: An Overview

Laboratory Medicine

• Laboratory medicine involves selecting, providing, and interpreting diagnostic tests using patient samples
• Testing can confirm or exclude diagnoses
• It assists in treatment selection, optimization, and monitoring
• It provides prognoses and screens for diseases without clinical signs or symptoms
• Testing establishes and monitors the severity of physiological disturbances

Disciplines of Laboratory Medicine

• Clinical biochemistry
• Histopathology
• Microbiology
• Cytology
• Blood banking
• Hematology
• Immunology
• Virology

Clinical Biochemistry

• Clinical biochemistry (chemical pathology) investigates bodily fluids like blood, urine, and cerebrospinal fluid
• Diseases are diagnosed and monitored by identifying changes in body chemistry
• Clinical biochemistry measures specific elements in biological samples
• These include proteins, sugars, cellular breakdown products, hormones, toxins, and electrolytes

Biochemical Tests

• Crucial to modern medicine
• Primarily use blood (plasma or serum), but may require urine, CSF, faeces, kidney stones, or pleural fluid
• Plasma is obtained by collecting blood with an anticoagulant and separating the liquid from cells via centrifugation
• Serum is the liquid obtained when blood is allowed to clot
• Plasma or serum use makes little difference in many blood tests

Types of Biochemical Investigations

• Urgent/STAT investigations are performed in emergencies
• Results are available from most labs within 24 hours, with a 1-hour turnaround
• Urgent tests include electrolytes (Na, K, Cl), blood gases (pO2, pCO2, pH, HCO3), oxygen saturation, cardiac markers, and glucose
• Routine investigations are a series of tests or diagnostic procedures is performed regularly to monitor health, with 3-hour turnaround time
• Special investigations are performed in larger centers with special chemistry labs due to specimen volume requirements and expense
• Special tests include hormones, specific proteins, vitamins, drugs, lipids, DNA analysis, and rare tests
• Organ panels are grouped investigations focusing on specific clinical presentations
• Examples of organ panels include liver function tests, kidney function tests, and lipid profiles

Collection of Specimens

• Test requests require unambiguous patient identification: name, sex, date of birth, and unique ID number
• They also require the location, requesting doctor's name, and sampling date/time
• Requests can be traditional forms with labeled specimens or electronic with a unique sample identifier (barcode)
• Each request must specify analyses, specimen details, and relevant clinical information
• Clinical labs require multiple procedures to avoid errors
• Errors can occur at different stages:
  • Pre-analytical: before measurement, involving sample collection errors (wrong tubes, timing, labels, tourniquets)
  • Analytical: laboratory errors like reagent contamination, pipetting, or computing mistakes (rare)
  • Post-analytical: decreasing due to electronic downloads, including transcription errors or misheard results

Quality Control

• Quality control in the clinical laboratory increases the probability of valid results used with confidence for diagnosis or therapy
• Achieved by ensuring accuracy, precision, and error-free reports

Accuracy versus Precision

• Accuracy refers to observations close to the "true" or "correct" value without systematic bias
• Lack of accuracy means that results will always tend to be either high or low
• Precision refers to reproducible or repeatable observations
• A precise method yields results close to one another (not necessarily the true value) on repeated analysis
• Lack of precision means that results may be scattered and unpredictably high or low

Descriptive Statistics

• High-volume clinical labs generate thousands of results daily
• The wealth of data must be summarized to monitor performance and is known as quality control
• Descriptive statistics provides the foundation to monitor performance

Descriptive Statistics - Measures of Center

• The three most commonly used descriptions of the center of a dataset are the mean, the median, and the mode
• The mean (average) is most commonly used
• The median is the "middle" data point, often used with skewed data
• The mode is rarely used to measure the center but describes data with two centers (bimodal)

Descriptive Statistics - Spread

• Describing data distribution (spread) is very useful
• The spread represents the relationship of all data points to the mean
• The range, standard deviation (SD), and coefficient of variation (CV) are commonly used to describe spread
• Standard deviation (s, SD, or σ) is the most frequently used measure of variation
• SD and variance indicate the "average" distance from the data center (mean) to each value; CV compares SDs with different units

Standard Deviation (SD)

• It measures "deviation" in terms of individual or grouped observations
• Deviation is measured by how far observations are from the mean
• Laboratories define "normal" or "acceptable" results within ± 2.0 SD from the mean
• Results exceeding 2.0 SD from the mean are "abnormal" or "out of control"
• This may indicate faulty laboratory testing

Example of Standard Deviation

• For establishing normal values, a minimum of 20 observations is needed
• Using a smaller sample size of 6 people to determine the normal range for fasting plasma glucose:
  • Glucose Levels (mg/dL): Ali 98, Ahmad 100, Nada 105, Sara 150, Noor 102, Tariq 101
  • Average = 109 mg/dL
  • SD = 20.0 mg/dL
  • 2 SD = 40.0 mg/dL
• The normal range for this group is 109 ± 40, or 69 − 149, which is ± 2.0 SD from the mean
• Sara's results are considered abnormal as her value is outside the normal range
• The Coefficient of variation (CV) for this group of observations is 18% of the mean

Control Charts

• QC systematically monitors analytic processes to detect errors and prevent incorrect patient results
• Monitoring involves assaying stable control materials and comparing determined values with expected values
• Expected values are represented by acceptable value intervals with control limits
• If expected values are within limits, the analytic method is properly reporting values
• If values fall outside limits, the operator is notified for further analysis before reporting patient results

The Control Charts

• Control samples are typically run daily, and values are plotted on Levy-Jennings charts
• Concentration is plotted on the Y-axis, and days are plotted on the X-axis
• A test is out of control if any deviation occurs beyond 2 SD in the charts

Example of Control Specimens

• If the Glucose result is 100 mg/dL, it may or may not be acceptable depending on the control specimen
• Assuming the established acceptable range is:
  • Mean = 104 mg/dL
  • SD = 5 mg/dL
  • 2.0 SD = 10 mg/dL
• The acceptable control range is 104 ± 10 = 94 - 104 mg/dL
• 94 – 104 mg/dl is considered acceptable dispersion; therefore, a control value of 100 mg/dL is within range

Coefficient of Variation

• The CV compares different sets of observations relative to their means
• Each SD reflects only the data that produced it - cannot be used for comparing groups
• The CV converts all observations into a percentage of relative means

Example of the usefulness of the CV

• Set A and Set B have different values of SD
  • Set A: SD=1.0, mean=10
  • Set B: SD=2.0, mean=1000
• Set B is more precise than Set A inspite of it having a greater value in the SD

Causes of Analytical Results Variation

• Inter-individual Variation: Age, Gender, Race, Genetics, Long-term health status
• Intra-individual Variation: Diet, Exercise, Drugs, Sleep pattern, Posture, Venipuncture time, Tourniquet application length
• Pre-analytical Variation: Transport Exposure to UV light, Standing time before separating cells, Centrifugation time, Storage conditions
• Analytical Variation: Random errors, Systematic errors
• Post-analytical: Transcription errors, results reported to the wrong patient

Reference Range

• The reference range (or interval) for a lab test determines whether a disease is present/absent or if the patient is at risk for future disease states
• It is used to monitor disease progression or therapeutic drug levels
• Reference ranges must be established or confirmed when:
  • A new analyte is measured
  • A new analytical method is introduced
  • There is a significant reagent modification by the manufacturer
• Labs are urged by manufacturers and required by CLIA to establish their own ranges

Reference Range Interpretation

• Results are typically interpreted with respect to value ranges found in normal, clinically healthy individuals
• A result is abnormal if it is less than the lower limit or greater than the upper limit of the normal range
• Reference ranges are determined from measurements on specimens from a large number (hundreds) of clinically normal individuals

Establishment of Reference Ranges

• Each lab must establish its own reference ranges
• Factors affecting reference ranges include:
  • Age, Sex, Diet, Medications
  • Physical activity, Pregnancy, Personal habits (smoking)
  • Geographic location (altitude), Body weight
  • Laboratory instrumentation (methodologies)
  • Laboratory reagents
• Ranges are defined as being within 2 Standard Deviations from the mean