Article Menu

View / Download pdf version of this article

In many scenarios, a diagnosis is based predominantly on the patient's history, with supporting evidence from physical exam and laboratory investigation. Diagnosis is rarely based on the results of laboratory investigations lonely, and testing is non always the best clinical form of action. Consideration needs to be given to whether testing volition add meaningful information to the overall clinical motion picture, and and so to the factors which may need to be taken into business relationship when interpreting the result. Can the test exist done immediately or is the patient required to prepare in some way, eastward.k. fasting? Is the patient taking any medicines that may influence or invalidate the results, e.thou. taking antibiotics prior to faecal antigen testing for Helicobacter pylori? Does the test need to be performed at a sure time, eastward.g. measuring testosterone in the morning? Is the test being washed at the right stage of affliction, e.grand. testing for antibodies subsequently seroconversion has occurred? A examination upshot returned from the laboratory will usually include a reference range or a threshold value based on guidelines. However, this result "on newspaper" does not e'er represent the clinical significance of a examination, which is simply apparent in one case all other factors for that individual patient have been taken into account.

This article is not intended to be a comprehensive guide, but rather an overview of general concepts to guide clinicians in considering the wide assortment of factors that can influence laboratory investigations or estimation of results.

If you are uncertain almost how a test may exist affected past a specific cistron you have identified, discuss this with the testing laboratory first, or include this detail on your investigation request grade.

To test or not?

Considerations prior to requesting a laboratory investigation include:

  • What is my reason for requesting this test? Due east.g. am I investigating symptoms and signs? Am I monitoring a disease or effect of a medicine? Is this a screening examination?
  • Has this test already been washed? Does it need to exist repeated? If then, when?
  • Will the test amend patient (or in some cases, family unit or partner) care?
  • Is this the correct examination or combination of tests for the clinical situation?
  • Is it the right fourth dimension to do the test?
  • How should the sample be taken?
  • How should the sample be stored and transported?
  • How will the test result exist interpreted?
  • How will the exam effect influence patient management?
  • What will be the consequences of a imitation positive upshot?
  • Are there potential harms of doing this test?
  • How will the patient be informed of the result?

Biological variation

There are certain variations in laboratory exam results that can be expected due to not-modifiable biological factors, such as age, biological rhythms and physiological changes during pregnancy. These factors may exist controlled for, e.chiliad. by selecting the most appropriate time in the day, month or year for a test, or may be taken into consideration in the interpretation of results, e.thousand. different reference ranges or thresholds for clinical significance depending on age, sexual practice or pregnancy status.

Advancing age

The physiological changes associated with ageing, forth with increasing co-morbidities and polypharmacy, mean that older people are more likely to accept exam results that autumn outside of the normal reference range. For some tests, laboratories are able to provide an age-adjusted reference range, just for other tests, a result outside of the range in an older patient, needs to be interpreted in the context of their overall clinical film.

In many cases, assessing the rate and magnitude of change over time offers more data than interpreting the value of an individual event. Often the population range of a test shows much more variation than that for an individual patient, e.g. serum creatinine, liver enzymes. In such cases the patient's ain previous results are a useful baseline.

An example of the effect of historic period, sex and other variables on interpretation of laboratory results is serum alkaline phosphatase (ALP), which may be requested as role of liver function tests. The upper reference limit is markedly increased during puberty as this is the time of maximum bone remodelling. After this period ALP levels autumn to a new upper limit through younger adult life, and then rise over again, peculiarly in females effectually the time of perimenopause, largely reflecting an increment in bone turnover at that time. Marked increases in serum ALP may also occur in women in late pregnancy (due to product of ALP by the placenta), and during other times such equally in the weeks after healing of a fracture.

Another example of laboratory values which change with age is lipids. In adults, total cholesterol, LDL and triglyceride levels increase with historic period, until approximately age l to sixty years in males and age 60 to seventy years in females, when they begin to reject in most people; triglyceride levels tend to continue to increment in older females.ane Many laboratories are no longer reporting reference ranges for lipid levels, every bit this is considered less clinically useful than treatment targets based on the underlying cause of raised levels and cardiovascular run a risk.two

When interpreting laboratory results in an older patient, ensure that an historic period-appropriate reference range is used if available. Normal age-related changes, e.g. deterioration in renal function, may explicate results outside of the reference range in an older patient, merely age lonely should not be considered as the only crusade of an abnormal outcome.

Biological rhythms

Many laboratory parameters vary depending on the time of mean solar day, week, month or year when they are sampled. Trunk temperature, hormone product (e.yard. cortisol, testosterone), platelet and cardiac function and cognitive function follow a circadian (24 hr) rhythm.1 To let for this effect, some laboratory tests are recommended at specific times of the twenty-four hours, due east.g. testosterone should be sampled betwixt 7 am – x am. This is because peak testosterone levels unremarkably occur in the early morning; evening levels are ofttimes substantially (upwardly to 50%) lower, especially in younger males.ii For most purposes, serum cortisol* should preferably be sampled in the early morning time equally there is marked diurnal variation, with early on morning levels at least 50 – 100% college than levels in the tardily afternoon.2

Testing vitamin D levels (25-hydroxyvitamin D) is seldom necessary, however, if levels are obtained, they need to be interpreted in the context of the time of year they are sampled. Seasonal variations occur with vitamin D, with the lowest levels unremarkably observed at the finish of winter (in countries with defined seasons, such as New Zealand). For example, if a patient has a mild vitamin D deficiency at the end of winter, this is likely to be less clinically significant than a patient with a mild deficiency at the terminate of summer.

For further data, encounter "Historic period-related testosterone decline in males" All-time Tests (Jun, 2012) and "Vitamin D supplementation: navigating the fence" BPJ 36 (Jun, 2011).

* Serum cortisol levels have a wide reference range, and therefore this is a relatively inaccurate measure of cortisol excess or deficiency. The dexamethasone suppression examination is used to assistance exclude Cushing's syndrome and the Synacthen stimulation test is used to investigate for primary or secondary hypoadrenalism.2

Menstrual cycle

Females who are menstruating take predictable monthly rhythms of FSH, LH, oestrogen and progesterone. In a general exercise setting, these hormone levels may be requested to investigate conditions such equally oligo/amenorrhoea or sub-fertility. The interpretation and meaningfulness of the results is dependent non only on the value, but on the stage in the wheel when the hormone was measured.

In a normal menstrual cycle LH levels pinnacle mid-cycle to trigger ovulation. FSH levels as well tiptop mid-wheel. Oestradiol (the predominant oestrogen in women who are ovulating) is highest prior to ovulation, and and so reduces if fertilisation does not occur. Unless ovulation is being investigated, levels of these hormones are best measured early in the menstrual cycle.2

N.B. Investigation of oestradiol levels is not useful in women who are taking oestrogen-containing oral contraceptives as this suppresses the pituitary ovarian axis. LH and FSH are too suppressed in women taking depot progesterone.two

Progesterone levels peak in the second phase of the menstrual cycle, subsequently ovulation has occurred, to prepare the endometrium for implantation of an embryo. Progesterone levels are sometimes measured to establish if ovulation has occurred; ordinarily 7 days before the expected date of menstruation, i.e. approximately day 21 if the woman has a regular 28 twenty-four hour period wheel.

Oestradiol levels decrease and FSH levels increment during menopause, but monitoring these hormone levels is not always reliable in predicting when a adult female is entering menopause, as fluctuations occur with varying ovarian activity.

For further information, see "Reproductive hormones: the correct test at the right time, for the right patient" All-time Tests (Feb, 2013).

Pregnancy

Physiological changes during pregnancy event in alterations in many laboratory parameters, such as blood volume, liver and renal role and hormone levels (Table 1, over page). Reference ranges for different stages of pregnancy are available for some laboratory tests, withal, these ranges are ofttimes not besides defined equally the general reference range. In improver, pregnancy-related changes, such as alterations in binding proteins, can bear upon assays differently, e.g. free hormone levels can be assay dependent. Therefore caution is recommended in interpreting results based on reference ranges and the laboratory should be contacted if there is any incertitude.

When requesting a laboratory investigation in a woman who is pregnant, notation the gestational week on the laboratory asking form.

Table i: Examples of laboratory values that change with pregnancy1, two, 3

Increases Decreases
Alpha-fetoprotein (AFP); peaks in third trimester Haemoglobin; due to haemodilution caused past greater claret volume
Alkaline phosphatase (ALP); up to four fold increase in third trimester Ferritin; decreases as pregnancy progresses
Blood volume (hateful plasma volume); increases by xxx–50% FT4; may decrease slowly in belatedly pregnancy (can be analysis dependent)
Lipids; up to 40% increment in cholesterol, triglyceride levels can markedly increase in some women (due to the consequence of oestrogen) Prothrombin and fractional thromboplastin times
Creatinine clearance; glomerular filtration rate increases 40-lx% (eGFR cannot be reliably calculated) TSH; decreases start trimester, then returns to normal (due to the effect of hCG)
ESR; increasing to 30–60 mm/h as pregnancy progresses Sodium; slight decrease due to changes in blood volume and fluid homeostasis
Hormones; oestrogen, testosterone, progesterone, human chorionic gonadotrophin (hCG), prolactin
Iron binding (transferrin levels); significant increase even in a non-fe deficient woman (due to the effect of oestrogen)
White blood count; may increase to 15–18 x109/L

Reference range

A reference range for a laboratory test is a statistically-derived numerical range of results that is obtained by testing a sample of "good for you" individuals. Defining "healthy", yet, is not straightforward, and depends on a wide range of factors and assumptions; in many cases to define a range simply using "perfectly healthy" patients would brand it unrealistic and unusable.

The range is too normally assumed to have a Gaussian distribution, in that 68% of values lie inside ane standard deviation (SD) of the mean value, 95% inside two SDs and 99.seven% of values within three SDs.1 However, many ranges practise non have a Gaussian distribution, simply rather the upper finish of the distribution is skewed. In these cases the range tin be derived either by log transformation of the data, or only by identifying the relevant 2.5th and 97.5th percentiles in the population being studied (with attempts to exclude those patients likely to have an underlying pathology by clinical, laboratory and statistical ways).

Reference ranges for laboratory results usually include two SDs from the hateful value significant that one in 20 "healthy" individuals will have a test result outside the reference range.i Reference ranges vary betwixt laboratories, and can change if new evidence becomes bachelor. The upper and lower limits of the range are not absolute and practise not define "normal" and "abnormal", but are points at which the probability of clinical significance tends to increase.

Some "reference ranges" are based on recommendations from international bodies for optimising patient outcomes, rather than on a population statistical distribution. For example, the upper reference limit for TSH in early on pregnancy is based on guidance statements from the Endocrine Lodge and the American Thyroid Association. The recommended limit for serum uric acid in patients taking uric acid lowering treatment (0.36 mmol/L) is based on the European League against Rheumatism (EULAR) guidelines.

Interpretation of a effect outside the stated reference range is therefore very dependent on the clinical background of the patient, the pattern of other abnormalities, and the clinical question(s) existence asked.

Individual variations

A patient's nutrition, wellness status and lifestyle factors can all have a pre-analytical influence on laboratory parameters.

Nutrition and nutritional condition

Fasting, calorie restriction, nutrient exclusion diets, malnutrition and dehydration tin all touch on laboratory results. The significance of some laboratory tests is dependent on decision-making dietary factors, east.g. ensuring there is sufficient gluten in the diet for at to the lowest degree several weeks prior to serology investigations for coeliac disease, or fasting prior to assessing the effect of intervention in a patient with previously loftier triglyceride levels. In other scenarios, assessing dietary factors tin help to interpret unexpected laboratory results, e.k. a vegetarian or vegan nutrition can result in decreased levels of vitamin B12, a low carbohydrate nutrition tin can cause increased ketone levels (every bit part of urinalysis) and a high poly peptide diet can result in increased uric acid levels.

Fasting for 12 hours prior to laboratory testing may exist helpful or even necessary, depending on the clinical scenario, to go the most authentic upshot for the following tests, which are affected past the ingestion of certain foods:4

  • Glucose; but if indicated – for almost patients HbA1c is now recommended equally the exam of showtime selection for the diagnosis and monitoring of type ii diabetes, and does not crave fasting
  • Triglycerides; for most patients fasting is not required for lipid testing, just may be useful for monitoring in people with high triglyceride levels
  • Uric acid; fasting is not usually required in do to become authentic results but the result of recent dietary intake may assist to explicate unexpected results
  • Creatinine; a recent repast with high meat content tin have a pregnant influence on serum creatinine, and this should be considered when monitoring eGFR5

Sustained low caloric intake and starvation can result in numerous changes to laboratory parameters such as glucose, thyroid function, electrolytes, liver function, renal function and lipids.4 Uric acid levels may be increased as a upshot of ketonaemia (causing reduced clearance).iii

Malnutrition has varying effects on laboratory results, depending on the nature of the patient's nutritional status. Malnutrition is classically idea of as a deficiency of protein and energy, with or without micronutrient deficiencies. However, malnutrition may be defined as under-nutrition, over-nutrition or deficiency of specific nutrients. Malnutrition should be considered equally a crusade for results such as decreased ferritin, folate and vitamin B12 levels.

Aridity tin can be considered as a crusade of sodium and potassium imbalances, and can as well affect numerous other indices, such every bit creatinine and urea, albumin, lipids and haematology indices.

For farther information see: "A primary care approach to sodium and potassium imbalance", Best Tests (Sep, 2011) and "Strategies to amend nutrition in elderly people", BPJ Special Edition (May, 2011).

Caffeine

The effect of caffeine on laboratory parameters has not been fully studied. It has a brusque half life of 3 to seven hours, but this varies among individuals.iv Caffeine intake causes transient increases in blood glucose levels and impairs glucose tolerance.3, 4 It can also impact other specialised investigations such as estimation of metanephrines when investigating hypertension.

Alcohol

The consequence of booze consumption on laboratory investigations depends on the duration and extent of use. Astute (transient) effects of alcohol consumption (within two to four hours) include decreased serum glucose and increased plasma lactate with a reduction in urinary uric acrid excretion due to the inhibition of hepatic gluconeogenesis.3

Chronic furnishings of booze consumption on laboratory investigations include:2

  • Elevated gamma glutamyl transferase (GGT) and mean jail cell book (MCV) which are commonly used to test for excessive alcohol consumption
  • Elevated aspartate aminotransferase (AST), alanine aminotransferase (ALT) and AST/ALT ratio
  • Elevated triglyceride levels
  • Elevated uric acid and ferritin levels due to fatty liver and alcoholic hepatitis
  • Elevated creatine kinase due to alcoholic myopathy
  • Other haematological abnormalities, e.g. anaemia and thrombocytopenia

Alcohol consumption tin likewise contribute to vitamin and mineral deficiencies due to replacement of nutrient with alcohol or as a upshot of interference of absorption of vitamin and minerals, e.g. decreased folate, vitamin A, vitamin B and calcium levels.

Timing of investigation in relation to stage of illness

The significance of an investigation tin can be dependent on when the sample was taken in relation to the stage of the disease procedure. The stage of illness can besides influence the pick of the nigh advisable investigation. For example, a serology test for syphilis may be falsely negative if the sample is taken likewise early after exposure, and therefore seroconversion has not yet occurred. Different types of serology test will provide data near active or past infection.

Acute illness tin also affect the result of some investigations, e.g. ferritin is an astute phase protein and levels can be increased past inflammation and infection, as well as chronic illness and malignancy.

Tobacco smoking

Regular smoking and exposure to nicotine tin can have both acute and chronic effects on laboratory investigations, although the mechanisms behind these changes are not fully understood. Within one 60 minutes of smoking one to five cigarettes, plasma/serum concentrations of fatty acids, adrenaline, glycerol, aldosterone and cortisol are increased.3 People who are chronic smokers may have persistent increases in leukocyte counts, heavy metals, lipoproteins, neoplasm markers and haematocrit (PCV), and decreases in the action of some enzymes (e.g. angiotensin-converting enzyme).three

Exercise

The effect of practice on laboratory parameters is dependent on the wellness status of the patient, air temperature during practice and intake of food and water during or following exercise.4 Extreme exercise or vigorous exercise in a person unaccustomed to this level of activity can result in changes to some laboratory parameters. For example, the most common cause of elevated creatine kinase (CK) levels is exercise. Intense exercise tin can crusade an acme in CK levels for several days to a calendar week. Well-muscled people often take CK levels persistently above normal.2

Thyroid part is also known to be altered in people undergoing loftier-intensity practise. For instance, anaerobic practise increases TSH and FT4 levels, just decreases FT3.4 Liver function (AST and to a lesser extent ALT) tests can increase after exercise. Transient proteinuria and haematuria are also mutual subsequently exercise, but usually resolve afterwards a few days. Other analytes that can exist increased by exercise include urea, creatinine, lactate dehydrogenase, prothrombin fourth dimension, and D-dimer levels. Fibrinogen and the activated fractional thromboplastin fourth dimension (APTT) tin can be reduced.iv Most of these furnishings are likely to be transitory (e.1000. persistent for a few hours to a few days afterward exercise), but this depends on individual patient factors.

Medicines

The medicines that a patient is taking can significantly impact some laboratory results, therefore this needs to be taken into consideration when interpreting results. Information technology is skilful practise to note the relevant medicines that a patient is taking on the laboratory request form, particularly if they may potentially influence results, e.g. antihypertensives existence taken when investigating secondary causes of hypertension or hormone replacement therapy being taken when requesting endocrine tests.

Medicines tin take a direct effect on the sample or laboratory testing process, causing an inaccurate event. For instance, when investigating for H. pylori, a fake-negative outcome of a faecal antigen test may occur in patients taking a course of antibiotics or proton pump inhibitors (PPIs) every bit this would decrease the gastric load of H. pylori.

Medicines may also cause a biological effect to the patient which would account for an altered upshot. For example, some antibiotics (due east.g. cotrimoxazole and erythromycin), cardiovascular medicines (eastward.grand. amiodarone and propranolol), NSAIDs (e.g. piroxicam) and gastrointestinal medicines (e.g. omeprazole) may business relationship for a raised INR result in a patient taking warfarin who normally has a stable INR.2 Long term use of metformin or PPIs is a possible explanation for a depression vitamin B12 level.2 Many medicines accept an result on the balance of sodium and potassium in the trunk, e.chiliad. diuretics may cause hypernatraemia (peculiarly loop diuretics), hyponatraemia (especially thiazides), hyperkalaemia (especially potassium-sparing diuretics) and hypokalaemia (loop and thiazide diuretics).2

For further information, see: "A primary care arroyo to sodium and potassium imbalance", Best Tests (Sep, 2011).

When monitoring the serum concentration or event of a medicine, the laboratory test needs to exist timed depending on the drug's metabolism, e.g. a blood sample for testing lithium levels should be collected 10–fourteen hours after the terminal dose and a sample for testing digoxin should exist collected at least eight hours after the terminal dose.2 When initiating a patient on warfarin, INR levels should be sampled daily in the morning, after an evening dose of warfarin, to calculate necessary dose adjustments.2

Belittling variation

Analytical variation occurs due to imperfections in testing methods and equipment, which may cause analyte values to be slightly dissimilar each fourth dimension they are measured. Modernistic testing methods and laboratory equipment mean that analytical variation is ordinarily less of a factor in differing test results than biological variation. Ideally the variation in measurement (expressed as analytical coefficient of variation, or CVa) should be less than half the individual patient biological variation of the analyte in question (CVi).

For further information on variation with specific analytes, see: www.westgard.com/biodatabase1.htm

Collection, storage and transport of samples

If a sample is being collected at the practise, it is important to be familiar with the blazon of collection container and sample medium that is required by the laboratory for the specific test, as this tin affect results, sometimes markedly. For example, a swab for PCR testing for pertussis should be transported in a dry tube or a tube with universal viral ship medium, just non in a tube with charcoal transport medium (which is acceptable for swab culture).

Other examples of drove or transport requirements for optimal test results include:

  • Blood samples for coagulation studies, including platelet count, D-dimer, prothrombin fourth dimension, APTT and fibrinogen, should exist transported to laboratory inside four hours of collectionsix
  • Urine specimens for civilisation should be stored in a refrigerator prior to transportation to reduce the rate of multiplication of microorganisms
  • Samples for glucose assay should be separated as before long as possible subsequently collection; this applies even with samples collected in fluoride or oxalate collection tubes, every bit reduction in glucose concentration nevertheless occurs for 60–ninety minutes
  • Samples for potassium or phosphate should not be left overnight, particularly in the fridge, as results tin be markedly altered, east.m. late evening drove with delayed send to the laboratory
  • Faecal samples for culture and microscopy should preferably be transported to the laboratory within 4 hours
  • Semen samples for fertility testing should be kept by the patient at body temperature, eastward.yard. by storing in a clothing pocket, and transported to the laboratory within one hr of collection.2 The same level of urgency is not required for post-vasectomy semen analysis.

Refer to your local laboratory for specimen drove requirements. Some providers have online resources, e.g. world wide web.labtests.co.nz/referrers/tests/collection-guide

Haemolysis

Haemolysis is the devastation of red claret cells, resulting in release of haemoglobin and cellular constituents, eastward.g. potassium, into the plasma. Information technology is a cause of inaccurate blood test results and can occur either in vitro during drove, storage or transportation of a blood sample or in vivo resulting in haemolytic anaemia if severe.

In vivo and in vitro haemolysis have different belittling features, e.1000. haptoglobin levels are normal with in vitro haemolysis, therefore analysis tin reveal if haemolysis occurred during the collection of the blood sample or if it was already present.2 The possibility of haemolysis should commonly be noted on results past the laboratory staff, including the likely caste of interference and reliability of the result. If the haemolysis is severe, a recollection of the sample may exist advised.

Some of the analytes that can exist affected when in vitro haemolysis has occurred include:2

  • Elevations in potassium, AST (ALT is less affected), lactate dehydrogenase, phosphate
  • Reductions in bilirubin, troponin T, insulin

Standardised sample collection and ship processes can assistance to forbid in vitro haemolysis, including:3

  • Allowing booze to dry completely when information technology is used for skin sterilisation prior to venepuncture
  • Not leaving a tourniquet on for longer than two minutes
  • Using an appropriately sized needle for collection (xx – 22 estimate needles can be used for most routine collections)
  • Non removing the needle from the vein if the vacuum tube is attached
  • Not exposing the specimen to extremes in temperature
  • Avoiding vigorous mixing or shaking of tubes
  • Avoiding delay in sending samples to the laboratory

Acknowledgement

Cheers to Dr Cam Kyle, Chemical Pathologist, Auckland for expert review of this article.

References

  1. Lee 1000, American Society of Health-Organization Pharmacists. Basic skills in interpreting laboratory data. Bethesda, MD: American Society of Health-Arrangement Pharmacists, 2013. Available from: www.123library.org/book_details/?id=108469 (Accessed Mar, 2015).
  2. Kyle C (Ed). Pathology handbook: a guide to the estimation of pathology tests. New South Wales: Sonic Healthcare, 2014.
  3. Guder WG, editor. Samples: from the patient to the laboratory: the affect of preanalytical variables on the quality of laboratory results. 3rd, rev ed. Weinheim, New York: Wiley-VCH, 2003.
  4. Peck Palmer OM. Result of historic period, gender, nutrition, exercise and ethnicity on laboratory exam results. In: Accurate results in the clinical laboratory: a guide to mistake detection and correction. London ; Waltham, MA: Elsevier, 2013. pp. 9–17.
  5. Priess D, Godber I, Lamb E, et al. The influence of a cooked meat repast on estimated glomerular filtration rate. Ann Clin Biochem 2007;44:35–42.
  6. Queensland Medical Laboratories. Pathology reference transmission. Available from: www.qml.com.au/Portals/0/PDF/RefManV2_WEB_APR09.pdf (Accessed Mar, 2015).