CLINICAL CHEMISTRY MARSHALL PDF
Clinical Chemistry, Seventh Edition - Marshall, William J. & Bangert, Stephen K & Lapsley, Marta - Ebook download as PDF File .pdf), Text File .txt) or read. Clinical Chemistry. 8th Edition. Authors: William Marshall Márta Lapsley Andrew Day. Paperback ISBN: eBook ISBN: In this issue of Clinical Chemistry, Mary Lopez and col- leagues (1) describe . Marshall J, Kupchak P, Zhu W, Yantha J, Vrees T, Furesz S, et al. Processing.
|Language:||English, Spanish, Japanese|
|Genre:||Academic & Education|
|ePub File Size:||21.62 MB|
|PDF File Size:||11.87 MB|
|Distribution:||Free* [*Regsitration Required]|
William J. Marshall, Stephen K. Bangert. Edinburgh ; New York: Mosby/Elsevier, pages, , English, Book; Illustrated, Clinical chemistry / William J. Clinical chemistry by William J Marshall · Clinical chemistry. by William J Marshall ; Marta Lapsley; Andrew Day. eBook: Document. English. Eighth edition. Köp Clinical Biochemistry E-Book av William J Marshall, Stephen K Bangert på wildlifeprotection.info PDF-böcker lämpar sig inte för läsning på små skärmar, t ex mobiler. From the author of the popular Clinical Chemistry medical student textbook.
However, although it may be tempting to perform all the assays on every specimen, this approach generates an enormous amount of information, some of which may be unwanted, ignored or misinterpreted e. Worst of all, it may actually divert the clinicians attention from important results. Discrete analysis, that is, performing only the necessary tests required to answer the clinical question e. Is this patients jaundice cholestatic or due to hepatocellular disease? Reporting results Once analysis has been completed and the necessary quality control checks made and found to be satisfactory, a report can be issued.
Cumulative reports, which show previous as well as current results, allow trends in data to be picked out at a glance.
It may be appropriate to add a comment to a report to assist the clinician with its interpretation. Results that indicate a need for rapid clinical intervention should be communicated to the requesting clinician as a matter of urgency. Point of care testing Not all analyses need to be performed in a central laboratory. Reagent sticks for testing urine at the bedside or in the clinic have long been available. Various substances, including glucose, protein, bilirubin, ketones and nitrites indicative of urinary tract infection , can be tested for using such sticks.
Testing of blood for analytes, such as glucose, and hydrogen ion and blood gases at point of care has also been available for some time. Indeed, the availability of easily used instruments to measure glucose allows patients with diabetes to monitor their blood glucose concentrations at home.
In recent years, manufacturers have developed instruments that can perform a wide range of tests suitable for use at the point of care. Such instruments allow the more rapid provision of analytical results for patients in whom they are required urgently e. It is clearly desirable that such instruments should be capable of providing results that are as robust with regard to accuracy and precision as those provided by the main laboratory.
These instruments are designed to be very simple to operate but it is nevertheless essential that individuals using them, who will usually not be laboratory staff, are properly trained in their use.
They should adhere to protocols designed to ensure the quality of results and to provide a robust audit trail so that, for example, should a manufacturer report a problem with a particular test, patients whose results may have been affected can be identified.
Both the training and quality issues should be supervised from the laboratory. Some analyses can be performed outside traditional healthcare settings and the results given directly to patients. An example is the measurement of plasma cholesterol concentration in retail pharmacies. Such analyses should be subject to appropriate quality assurance procedures and trained personnel should be available to advise patients on the significance of the results.
Sources of error Erroneous results are at best a nuisance; at worst, they have potential for causing considerable harm. Errors can be minimized by scrupulous adherence to robust, agreed protocols at every stage of the testing process: this means a lot more than ensuring that the analysis is performed correctly. Errors can occur at various stages in the process: pre-analytical, occurring outside the laboratory e.
Analytical errors can be systematic also known as bias: different analytical methods may produce results that are higher or lower it is to be hoped only slightly so than the definitive or reference method or random. Many of the few errors that do occur even in good laboratories are detected by quality control procedures, including data-handling software or personal scrutiny of reports by laboratory staff.
Some are so bizarre that they are easily recognized for what they are. More subtle ones are more likely to go undetected. Unfortunately, the risk of errors occurring can never be entirely eliminated. Interpretation of results When the result of a biochemical test is obtained, the following points must be taken into consideration: Is it significantly different from any previous results? Is it consistent with the clinical findings?
Is it normal? The use of the word normal is fraught with difficulty. Statistically, it refers to a distribution of values from repeated measurement of the same quantity and is described by the bell-shaped Gaussian curve Fig. Many biological variables show a Gaussian distribution: the majority of individuals within a population will have a value approximating to the mean for the whole, and the frequency with which any value occurs decreases with increasing distance from the mean.
The range of the mean 2 standard deviations SDs encompasses The range of the mean 3 SDs encompasses For some analytes, the distribution of values is skewed; an example is plasma bilirubin concentration. Such data can often be mathematically transformed to a normal distribution: data distributed with a skew to the right of the mean as is the case with bilirubin can often be transformed to a normal distribution if re-plotted on a semi-logarithmic scale.
When establishing the range of values for a particular variable in healthy people, it is conventional to first examine a representative sample of sufficient size to determine whether or not the values fall in a Gaussian distribution. The range mean 2 SDs can then be calculated; this, in statistical terms, is the normal range. This suggests that, if the measurements were to be made in a group of comparable individuals, 1 in 20 would have a value outside this range.
The specialized statistical use of the word normal does not equate with what is generally meant by the word, that is, habitual or usually encountered. The statistical normal may not be related to another common use of the word, which is to imply freedom from risk. For example, there is an association between increased risk of coronary heart disease and plasma cholesterol concentrations even within the normal range as derived from measurements on apparently healthy men.
Thus, the normal range for an analyte, defined and calculated as described, has severe limitations. It only identifies the range of values that can be expected to occur most often in individuals who are comparable with those in the population for whom the range was derived.
It is not necessarily normal in terms of being ideal, nor is it associated with no risk of having or developing disease. Furthermore, by definition it will exclude values from some healthy individuals.
In all cases, like must be compared with like. When physiological factors affect the concentration of an analyte see Fig. It may, therefore, be necessary to establish normal ranges for subsets of the population, such as various age groups, or males or females only. To alleviate the problems associated with the use of the word normal, the term reference interval RI often called the reference range has been widely adopted by laboratory staff, using numerical values reference limits generally based on the mean 2 SDs.
Results can be compared with the RI without assumptions being made about the meaning of normal.
In practice, the term normal range is still in general use outside laboratories. It is used synonymously with reference interval in this book. Reference intervals for some common analytes are given in the Appendix: these are as used in one of the authors laboratories, and are appropriate for the case histories, but may not apply to other laboratories because of differences in analytical methods and in the characteristics of the population on which the data are based.
Differences between reference ranges are a particular problem with immunoassays, since different antibodies may vary in their specificity for the analyte and the extent to which they exhibit cross-reactivity with other, similar molecular species.
Nevertheless, efforts are being made in the UK to introduce common standards in various areas of pathology, including uniform reference ranges. In using RIs to assess the significance of a particular result, the individual is being compared with a population. Some analytes show considerable biological variation, but the combined analytical and biological variations will usually be less for an individual than for a population. Thus, it is possible for a test result to be abnormal for an individual, yet still be within the accepted normal range.
An abnormal result does not always indicate the presence of a pathological process, nor a normal result its absence. However, the more abnormal a result, that is, the greater its difference from the limits of the reference interval, the greater is the probability that it is related to a pathological process.
In practice, there is rarely an absolute demarcation between normal values and those seen in disease: equivocal results must be investigated further. If an important decision in the management of a patient is to be based upon a single result, it is vital that the cut-off point, or decision level, is chosen to ensure that the test functions efficiently.
In screening for PKU, for example, the blood concentration of phenylalanine selected to indicate a positive result must include all infants with the condition; in other words, there must be no false negatives. Because there is some overlap in the values seen in the presence and absence of PKU, this inevitably means that some normal children will test positive false positives and will be subjected to further investigation.
Generally, it is unusual to have to determine a patients management on the basis of one result alone. If a second and independent variable is measured, the probability that this result will be abnormal is also 0.
It follows that the more tests that are performed on an individual, the greater the probability that the result of one of them will be abnormal: for 10 independent variables, the probability is 0. For 20 variables, the probability is 0. Although biochemical parameters are frequently, to some extent, interdependent e. Before any decision can be made on the basis of such results, some information is required about the probability that they are indicative of a pathological process.
This topic is discussed on p. Is it different? If the result of a previous test is available, the clinician will be able to compare the results and decide whether any difference between them is significant.
This will depend on the precision of the assay itself a measure of its reproducibility and the natural biological variation.
Illustrated textbook of clinical chemistry
Some examples of variation in common analytes are given in Figure 1. Analytical variation: typical standard deviations for repeated measurements made using a multichannel analyser on a single quality control serum with concentrations in the normal range. Biological variation: means of standard deviations for repeated measurements made at weekly intervals in a group of healthy subjects over a period of ten weeks, corrected for analytical variation.
Thus, for plasma calcium concentration, with an analytical SD of 0. However, to decide whether an analytical change is clinically significant it is necessary to consider the extent of natural biological variation. The effects of analytical and biological variation can be assessed by calculating the overall standard deviation of the test, given by: where SDA and SDB are the SDs for the analytical and biological variation, respectively.
If the difference between two test results exceeds 2. It does not mean that a difference of less than that equating to this probability cannot be of significance, nor that a greater difference necessarily is significant. If undertaking a major intervention depends on a result, it may be desirable only to make this decision if the probability that the change is not the result of innate variation is considerably greater.
Case history 1. Six months later, both conditions were well controlled and the test was repeated. Comment The analytical variation for creatinine is 5.
Is it consistent with clinical findings?
If the result is consistent with clinical findings, it is evidence in favour of the clinical diagnosis. If it is not consistent, the explanation must be sought.
There may have been a mistake in the collection, labelling or analysis of the sample, or in the reporting of the result. In practice, it may be simplest to request a further sample and to repeat the test. If the result is confirmed, the utility of the test in the clinical context should be considered and the clinical diagnosis itself may have to be reviewed. The clinical utility of laboratory investigations In using the result of a test, it is important to know how reliable the test is and how suitable it is for its intended purpose.
Thus, the laboratory personnel must ensure, as far as is practicable, that the data are accurate and precise, and the clinician should appreciate how useful the test is in the context in which it is used. Various properties of a test can be calculated to provide this information.
Specificity and sensitivity Earlier in the chapter, the terms sensitivity and specificity were used to describe characteristics of analytical methods. The terms are also widely used in the context of the utility of laboratory tests. The specificity of a test is a measure of the incidence of negative results in persons known to be free of a disease, that is, true negative TN.
Sensitivity is a measure of the incidence of positive results in patients known to have a condition, that is, true positive TP.
Because the ranges of results in quantitative tests that can occur in health and in disease almost always show some overlap, individual tests do not achieve such high standards. Factors that increase the specificity of a test tend to decrease the sensitivity, and vice versa. On the other hand, the test would have a low sensitivity in that many patients with mild hyperthyroidism would be misdiagnosed.
These concepts are illustrated in Figure 1.
Courses, Curricula, and Faculty
If the diagnostic cutoff value for a test is set too high B , there will be no false positives, but many false negatives; specificity is increased but sensitivity decreases. If the diagnostic cut-off value is set too low C , the number of false positives, and sensitivity, increases, at the expense of a decrease in specificity.
Whether it is desirable to maximize specificity or sensitivity depends on the nature of the condition that the test is used to diagnose and the consequences of making an incorrect diagnosis.
For example, sensitivity is paramount in a screening test for a harmful condition, but the inevitable false positive results mean that all positive results will have to be investigated further. However, in selecting patients for a trial of a new treatment, a highly specific test is more appropriate to ensure that the treatment is being given only to patients who have a particular condition. In some cases, this decision may not be straightforward, for example in the context of chest pain and suspected acute myocardial infarction, where the possible options are to identify all those who have had a myocardial infarction rule in or to identify all those who have definitely not rule out.
The preferred option should depend on the relative outcomes of treatment and non-treatment for patients in the two groups. One way of comparing the sensitivity and specificity of different tests is to construct receiver operating characteristic curves ROC curves. Each test is performed in each of a series of appropriate individuals. The specificity and sensitivity are calculated using different cut-off values to determine whether a given result is positive or negative Fig.
The curves can then be assessed to determine which test performs best in the specific circumstances for which it is required. Examination of the curves shows that test A performs less well in terms of both sensitivity and specificity than tests B and C. Test B has better specificity than C, but C has better sensitivity. The specialized use of the terms sensitivity and specificity that has been discussed here in the context of the utility of laboratory tests sometimes causes confusion, as these terms are also used to describe purely analytical properties of tests.
Readers should appreciate that, in this latter context, sensitivity relates to the ability of a test to detect low concentrations of an analyte and specificity to its ability to measure the analyte of interest and not some other usually similar substance. Efficiency The efficiency of a test is the number of correct results divided by the total number of tests. Thus efficiency is given by: When sensitivity and specificity are equally important, the test with the greatest efficiency should be used.
Predictive values Even a highly specific and sensitive test may not necessarily perform well in a clinical context. This is because the ability of a test to diagnose disease depends on the prevalence of the condition in the population being studied prevalence is the number of people with the condition in relation to the population.
This ability is given by the predictive value PV. The approach and scope of this trusted text makes it ideal for integrated medical curricula for medical training and for students and practitioners of clinical and biomedical science. The complementary online version of the book, including additional self-assessment material, completes this superb learning package.
Comes with complete, downloadable eBook on Student Consult. New introductory chapter on basic biochemistry - including solutions, solutes, ionisation, pH, buffers, amino acids, peptides and proteins, enzyme activity, including kinetic properties, DNA structure. Updated references to core guidelines UK and international reflect latest best practice.
Consultant Chemical Pathologist, Epsom and St. We are always looking for ways to improve customer experience on Elsevier. We would like to ask you for a moment of your time to fill in a short questionnaire, at the end of your visit. If you decide to participate, a new browser tab will open so you can complete the survey after you have completed your visit to this website. Thanks in advance for your time.
Skip to content.
Clinical Chemistry, Seventh Edition - Marshall, William J. & Bangert, Stephen K & Lapsley, Marta
Search for books, journals or webpages All Webpages Books Journals. Paperback ISBN: Published Date: Page Count: Instructor Ancillary Support Materials. Free Shipping Free global shipping No minimum order.
An Introduction to Biochemistry and Cell Biology 2. Biochemical Investigations in Clinical Medicine 3. Water, Sodium and Potassium 4.When sodium overload is due to excessive intake. Acidification of the urine takes place primarily in the distal parts of the distal convoluted tubules and in the collecting ducts. Ideally suited for preparation for the MRCPath and similar examination. Loss of bicarbonate Loss of bicarbonate and retention of hydrogen ions can result in acidosis in patients losing alkaline secretions from the small intestine e.
It measures low concentrations of the analyte sensitive and is not subject to interference by other substances specific. However, it remains the case that many well-established tests have been introduced into clinical practice without being properly evaluated, and few systematic reviews of existing tests have been performed.