Esa Leppänen

Effects of psychological stress on plasma interleukins-1 beta and 6, C-reactive protein, tumour necrosis factor alpha, anti-diuretic hormone and serum Cortisol

The study was undertaken to determine whether psychological stress influences immunobiological functions and is an important preanalytical factor to be considered in connection with blood specimen collection. Two kinds of stress were applied, the Stroop colour conflict test and the thrill of a novice about to make the first jump with a parachute. In both test situations, the level of the stress indicators cortisol or anti-diuretic hormone rose significantly. The concentrations of the cytokines studied did not change significantly. However, in the parachute test significant positive correlations were found, e.g. between the changes of cortisol and C-reactive protein and between anti-diuretic hormone and interleukin-1 beta. This suggests that there is an interaction between the endocrine and the immune systems in the response to a psychological stress.

Preanalytical factors and the measurement of cytokines in human subjects

Cytokines are widely measured in research. However, cytokine analyses are influenced by a myriad of factors. For instance, a delay in the separation of plasma from cells may lead to a 50% decrease in the concentration of tumor necrosis factor in plasma. Another example is the secretion of interleukin-1β in women which can be twice as high during the follicular phase as in the luteal phase. The factors influencing the outcome of these tests can be divided into in vivo preanalytical factors (e.g., aging chronobiological rhythms, diet, etc), in vitro preanalytical factors (e.g., specimen collection, equipment, transport, storage, etc), and analytical factors. To improve the value of the cytokine tests, factors strongly influencing the results have to be controlled. This can be done by using standardized assays and specimen collection procedures. In general, sufficient attention is not given to the preanalytical factors, especially in the measurement of cytokines. This article reviews the preanalytical factors which may influence the outcome of these tests in human subjects.

Short-Term Variability in the Concentration of Serum Interleukin-6 and Its Soluble Receptor in Subjectively Healthy Persons

Abstract The effects of the time of the day and breakfast on the outcome of interleukin-6 and its soluble receptor in serum were studied on 22 subjectively healthy volunteers. We collected blood specimens at 8.00 a.m. after an overnight fast, at 9.30 a.m. after usual breakfast, and at 11.30 a.m. A second group of volunteers (n = 20) was studied during the evening. We observed a significant decrease in the concentration of interleukin-6 after breakfast. This suggests that for assay of this analyte it is preferable to collect specimens after overnight fast. No changes were observed in the concentrations of interleukin-6 soluble receptor. Weak, but significant negative correlations were found between the age of the volunteers and the concentrations of interleukin-6 soluble receptor and between age and the ratio interleukin-6 soluble receptor/interleukin-6. The concentrations of interleukin-6 were positively and significantly correlated with age. The within-subject variablity was 30 % for interleukin-6,13% for its receptor, and 33.1% for the ratio interleukin-6 soluble receptor/interleukin-6. The between-subject variability was 43.7 % for interleukin-6, 22.2% for its receptor, and 60.2 % for the ratio interleukin-6 soluble receptor/interleukin-6. The indices of individuality (ratio within-subject variability/between-subject variability) were calculated and found to be 0.69, 0.60, and 0.55 for interleukin-6, its soluble receptor, and the ratio interleukin-6 soluble receptor/interleukin-6, respectively. The implications of these indices with the use of health-related reference values is discussed.

Preanalytical factors and standardized specimen collection: Influence of psychological stress

In order to devise and evaluate standardized specimen collection procedures, we studied the influence of psychological stress on the results of commonly analysed blood components: creatine kinase, lactate dehydrogenase, total protein and albumin in serum and blood picture. In addition, serum cortisol was assayed. Two kinds of stress were used: the Stroop test, a colour conflict task, and the thrill caused by the first jump of new parachutists. More changes were observed after the parachutist test than after the Stroop test. There was a difference in the responses of males and females. Females were more sensitive, especially to the parachutist test. Most of the changes observed were interpreted as being caused by haemoconcentration, possibly related to muscular tension. Cortisol, commonly used to indicate the level of stress, did not react much and is therefore not a good index of psychological stress.

Experimental basis of standardized specimen collection: the effect of short moderate exercise on serum K, Na, ASAT, ALAT, CK and LD.

Strenuous exercise is known to produce marked changes in many serum components. Present recommendations concerning drawing of blood for laboratory tests, therefore, allow little exercise before specimen collection. However, the effects of moderate exercise are unknown. Moderate exercise administered by bicycle ergometer (120-160 W for 15 min) to healthy persons and ambulant patients (average 107 W) induced a rapid and statistically significant effect on serum potassium, sodium, ASAT, CK and LD concentrations. However, the observed changes were largely abolished by having the patient to sit for 15 min before venipuncture. It appears to be possible to allow moderate exercise before venipuncture as long as the specimen collection procedure itself is sufficiently standardized, i.e. by specifying sitting for 15 min and no use of tourniquet.

Are the Preanalytical Factors Underestimated in Clinical Studies

Sir, In their recent review article dealing with the intraand inter-individual biological variability (1), SebastiánGámbaro et al. state that the variation which occurs in the results of specimen analyses consists of two components, metrological and biological. Moreover, in more than 25 % of the examples displayed, the intra-individual variation was larger than the inter-individual variation. The majority of the analyses were performed on urine. Different classifications of the factors influencing the results exist. We have found it practical to divide these factors into preanalytical and analytical (2, 3). Preanalytical factors occur before the final analysis and may be divided into in vivo (or biological) and in vitro factors. The in vivo factors occur in the subject before and during the specimen collection. Some may be controllable (e.g. diet, drug regimen, physical activity, posture) and some not (e.g. genetic factors, gender, age). The in vitro factors include the collection (equipment) and processing of the specimen (transport, storage, preparation for the analysis). The intra-individual variation in the concentration of analytes has often been observed to be larger than the variation between individuals. Though well documented (4), this phenomenon is unexpected, almost illogical. The following exaggerated example may elucidate the cause of such findings. A volunteer runs a Marathon run (laboratory results from such enthusiasts have been published) (5), afterwards he jumps into icecold water (such studies have also been conducted) (6). We would probably observe considerable variation between the data obtained in such situations. In a typical population study, volunteers or patients visit a clinic at a suitable time of the day (usually before or after work). They wait for a while (probably sitting) and then give blood. It would be hardly surprising if the intra-individual variation were not greater in the athlete of the first example. In the second example the preanalytical conditions would be fairly similar for all individuals. Our conclusion is that one should try to work out a convention regarding the production of data on intraindividual variation in studies aiming at generation of data to be used for interpreting results from patients. Though recommendations on the generation and application of data on biological variation have already been published (e.g. 7), the conditions under which the biological variability is studied should be clearly defined. We suspect that had the preanalytical factors been better controlled, in many reports the intra-individual biological variability would have been lower.

Experimental basis of standardized specimen collection: the effect of moderate ethanol consumption on some serum components (K, Na, ASAT, ALAT, CK, LD, total protein)

Venous blood was collected on two mornings from seven healthy volunteers using the standard Scandinavian procedure (fasting, sitting and no tourniquet) and serum Na, K, ASAT, ALAT, CK, LD and total protein were assayed. Then ethanol (0.75 g/kg body weight) was given on three consecutive evenings and the subsequent observed values 1, 3, 15, 38, 62 and 110 h post-ethanol were compared with the morning values. The mean component/total protein ratios dropped 14% for ASAT and 19% for ALAT 62 and 15 h post-ethanol, respectively. CK rose 17% at 3 h and dropped 17% at 62 h. However, the absolute values of ASAT, ALAT and CK did not change significantly. The only significant post-ethanol changes occurred in Na +2.14 mmol/l at 1 h (p less than 0.01), K -0.24 mmol/l at 3 h (p less than 0.05) and +0.26 mmol/l at 15 h (p less than 0.05), and LD +31 IU/l at 15 h (p less than 0.05) and +25 IU/l at 110 h (p less than 0.01). In one series, the total protein concentration dropped 4.57 g/l at 110 h post ethanol (p less than 0.001) but this drop was not reproducible in two additional series. It is concluded that moderate (social) ethanol consumption does not produce clinically significant effects on the components analysed. Also it may be misleading to express results as component/total protein ratios.

The effect of the order of filling tubes after venipuncture on serum potassium, total protein, and aspartate and alanine aminotransferases

The serum components mentioned were measured from the first and last tubes filled with blood following one venipuncture performed without use of a tourniquet on ordinary outpatients. The concentration of potassium was significantly higher in the first tube, whereas the concentration of the other components did not change. In a second series of experiments the first drops of blood (approximately 1 ml) emerging from the needle were collected separately. In the serum from this tube, the potassium concentration was significantly higher than in the serum from the second tube (approximately 10 ml), whereas the sodium concentration was the same. It is concluded that serum potassium should routinely be assayed from the second or a later tube of blood filled after one venipuncture. The S-ASAT, S-ALAT and S-total protein may be assayed from any tube.