| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department KKA, Clinical Biochemistry, Odense University Hospital, Odense, Denmark
aAddress correspondence to this author at: Department KKA, Clinical Biochemistry, Odense University Hospital, DK-5000 Odense C, Denmark. Fax 45-6541-1911; e-mail mads.nybo{at}ouh.fyns-amt.dk.
To the Editor:
For more accurate reporting of studies of diagnostic accuracy, the STARD (Standards for Reporting of Diagnostic Accuracy) steering committee has published a checklist and a generic flow diagram [see, for an example, the checklist in Clinical Chemistry (1)]. The intention is to improve the quality and evaluation of research to the advantage of clinicians, researchers, reviewers, journals, and the public. In clinical biochemistry, many diagnostic aspects need standardization. Such standardization unfortunately does not exist for the term "fasting", which may imply not only the duration of the fasting period but also restrictions concerning smoking, alcohol, coffee/tea consumption, and physical activity.
When the patient is sent to the laboratory for fasting blood sampling, fasting instructions are given by the referring doctor. If the definition of fasting is not standardized, patients will receive various instructions, and the diagnostic value of the analysis, which is particularly important for diagnosis of lipid disorders and diabetes, will be questionable. Therefore, an accurate definition of fasting is required.
We made a systematic literature search, using Medline 1965–2004 and PubMed (National Library of Medicine, NIH), to elucidate the effect of fasting on biochemical indices of diagnostic relevance. Titles and abstracts were thoroughly examined, and 835 articles were chosen for critical appraisal. Specific search strategies and search terms can be obtained from the corresponding author. Evaluation of the literature revealed solid evidence for the need of fasting, because different aspects [e.g., duration of fasting (2), smoking (3), coffee(4), and physical activity (5)] affect important biochemical indices.
On the basis of these findings, we conducted a questionnaire study concerning different issues within the fasting definition (i.e., fasting duration and restrictions on smoking, alcohol, coffee/tea, or physical activity). We sent this questionnaire to university hospitals and larger departments of clinical biochemistry in Denmark, making sure that all counties were represented to reveal "local" definitions of fasting. Eighty percent of the departments answered the questionnaire, and the answers revealed substantial variety in the fasting definitions used: Duration of fasting was not consistently defined, and 12.5% of the departments had no well-defined duration of the fasting period at all; 25% of the departments did not mention the issues of smoking, tea/coffee, or physical activity in their fasting criteria.
We also examined articles published in 2002 in Clinical Chemistry, Clinical Chemistry and Laboratory Medicine, Diabetes, and the Scandinavian Journal of Clinical and Laboratory Investigation to determine whether articles reporting the use of blood samples drawn from fasting individuals presented a proper fasting definition in a scientific context. As Table 1
shows, the percentage of articles with a clear fasting definition ranged from 0% to 17% in the journals examined. Fasting was considered well defined when, as a minimum, the definition comprised the duration of fasting, the amount of liquid intake allowed, and whether smoking was permitted. The conventional term "overnight fast" was not accepted as being well defined. When fasting durations were specified, we found considerable differences (8 to 14 h), which underscores the need for inclusion of such information in scientific reports.
|
Our investigations show a considerable lack of a uniform fasting definition containing all of the aspects addressed here. The lack is pronounced in local milieus as well as in the scientific literature. A standardized and evidence-based definition of fasting is therefore needed. Indeed, our findings indicate that, as a part of the STARD initiative (1), any clinical study should contain a comprehensive definition of fasting that includes all aspects important for the issue being addressed.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
