Requesting Laboratory Tests and Interpreting the Results

Requesting Laboratory Tests and Interpreting the Results


There are many laboratory tests available to the clinician. Correctly used, these may provide useful information, but, if used inappropriately, they are at best useless and at worst misleading and dangerous.

In general, laboratory investigations are used:

  • to help diagnosis or, when indicated, to screen for metabolic disease,

  • to monitor treatment or detect complications,

  • occasionally for medicolegal reasons or, with due permission from the patient, for research.

Overinvestigation of the patient may be harmful, causing unnecessary discomfort or inconvenience, delaying treatment or using resources that might be more usefully spent on other aspects of patient care. Before requesting an investigation, clinicians should consider whether its result would influence their clinical management of the patient.

Close liaison with laboratory staff is essential; they may be able to help determine the best and quickest procedure for investigation, interpret results and discover reasons for anomalous findings.


This depends on the following:

  • How quickly numerically significant changes are likely to occur: for example, concentrations of the main plasma protein fractions are unlikely to change significantly in less than a week (see Chapter 19), similarly for plasma thyroid-stimulating hormone (TSH; see Chapter 11). See also Chapter 3.

  • Whether a change, even if numerically significant, will alter treatment: for example, plasma transaminase activities may alter within 24 h in the course of acute hepatitis, but, once the diagnosis has been made, this is unlikely to affect treatment (see Chapter 17). By contrast, plasma potassium concentrations may alter rapidly in patients given large doses of diuretics and these alterations may indicate the need to instigate or change treatment (see Chapter 5).

Laboratory investigations are very rarely needed more than once daily, except in some patients receiving intensive therapy. If they are, only those that are essential should be repeated.


The main reason for asking for an investigation to be performed ‘urgently’ is that an early answer will alter the patient’s clinical management. This is rarely the case and laboratory staff should be consulted and the sample ‘flagged’ as clearly urgent if the test is required immediately. Laboratories often use large analysers capable of assaying hundreds of samples per day (Fig. 1.1). Point-of-care testing can shorten result turnaround time and is discussed in Chapter 30.

Figure 1.1 A laboratory analyser used to assay hundreds of blood samples in a day. Reproduced with kind permission of Radiometer Limited.

Laboratories usually have ‘panic limits’, when highly abnormal test results indicate a potentially life-threatening condition that necessitates contacting the relevant medical staff immediately. To do so, laboratory staff must have accurate information about the location of the patient and the person to notify.


When interpreting laboratory results, the clinician should ask the following questions:

  • Is the result the correct one for the patient?

  • Does the result fit with the clinical findings?

  • Remember to treat the patient and not the ‘laboratory numbers’.

  • If it is the first time the test has been performed on this patient, is the result normal when the appropriate reference range is taken into account?

  • If the result is abnormal, is the abnormality of diagnostic significance or is it a non-specific finding?

  • If it is one of a series of results, has there been a change and, if so, is this change clinically significant?

Abnormal results, particularly if unexpected and indicating the need for clinical intervention, are best repeated.

Test reference ranges

By convention, a reference (‘normal’) range (or interval) usually includes 95 per cent of the test results obtained from a healthy and sometimes age- and sex-defined population. For the majority of tests, the individual’s results for any constituent are distributed around this mean in a ‘normal’ (Gaussian) distribution, the 95 per cent limits being about two standard deviations from the mean. For other tests, the reference distribution may be skewed, either to the right or to the left, around the population median. Remember that 2.5 per cent of the results at either end will be outside the reference range; such results are not necessarily abnormal for that individual. All that can be said with certainty is that the probability that a result is abnormal increases the further it is from the mean or median until, eventually, this probability approaches 100 per cent. Furthermore, a normal result does not necessarily exclude the disease that is being sought; a test result within the population reference range may be abnormal for that individual.

Very few biochemical tests clearly separate a ‘normal’ population from an ‘abnormal’ population. For most there is a range of values in which ‘normal’ and ‘abnormal’ overlap (Fig. 1.2), the extent of the overlap differing for individual tests. There is a 5 per cent chance that one result will fall outside the reference range, and with 20 tests a 64 per cent chance, i.e. the more tests done, the more likely it is that one will be statistically abnormal.

No result of any investigation should be interpreted without consulting the reference range issued by the
laboratory carrying out the assay. Some analytes have risk limits for treatment, such as plasma glucose (see Chapter 12), or target or therapeutic limits, such as plasma cholesterol (see Chapter 13).

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 5, 2016 | Posted by in BIOCHEMISTRY | Comments Off on Requesting Laboratory Tests and Interpreting the Results

Full access? Get Clinical Tree

Get Clinical Tree app for offline access