delta-Aminolaevulinic acid in plasma, cerebrospinal fluid, saliva and erythrocytes: studies in normal, uraemic and porphyric subjects

delta-Aminolaevulinic acid (ALA) was determined by g.l.c. with electron-capture detection. Normal plasma level was 92 nmol/l (SD = 39, n = 89, range 24-270 nmol/l). ALA was undetectable in 35 of 53 samples of normal cerebrospinal fluid (limit of assay 2 nmol/l). The mean of the other 18 samples was 19 nmol/l (SD = 10, range 6-36 nmol/l). Salivary ALA was generally only 10-30% of the plasma level in normal and porphyric subjects. Erythrocytes of normal and porphyric subjects contained no detectable ALA and were impermeable to its entry. ALA clearance correlated closely with that of creatinine, consistent with it being excreted by glomerular filtration with limited tubular reabsorption. In chronic renal failure, serum ALA was elevated to a maximum of three to four times the normal, but its urinary excretion was reduced, in keeping with lessened production. In two cases of acute intermittent porphyria with overwhelming neuropathy the maximum plasma levels of ALA were 9 and 12 mumol/l. Haematin infusion decreased the ALA levels but without obvious clinical benefit. Limited neurological recovery occurred without major reduction in plasma levels of ALA. One subject's attack was precipitated by pregnancy. The neonate was apparently normal, despite high levels of ALA in maternal plasma throughout gestation and a high level of ALA in the cord blood. The observations described here do not support the view that ALA may be directly neurotoxic.     

Determination of plasma oxalate with oxalate oxidase

A method for the determination of plasma oxalate using oxalate oxidase (EC 1.2.3.4) and deproteinised plasma is described. Reference values are 1.2-6.4 mumol/l (n = 24, mean +/- SD 3.3 +/- 1.5 mumol/l). The sensitivity is 6-7 nmol, the accuracy 3-5 nmol, and the coefficient of variation 10.4% (at a level of 23 nmol). The recovery from plasma spiked with oxalate was 105 +/- 8% (n = 8). Allowing blood to stand for 4 h at room temperature had no effect on plasma oxalate levels; inhibitors of glyoxylate converting enzymes interfered with oxalate oxidase.     

Increase in blood plasma levels of boron and strontium in hemodialyzed patients

Boron and strontium concentrations in blood plasma of controls and hemodialyzed patients from two Centers were determined by inductively coupled plasma emission spectrometry. Boron concentrations in blood plasma were respectively, in controls 2.6 +/- 0.9 mumol/l and in hemodialyzed patients 16.1 +/- 5.6 mumol/l before the dialysis session and 9.5 +/- 3.2 mumol/l at the end. The decrease in blood plasma during the dialysis was concomitant with an increase in the dialysis fluid (1.2 +/- 0.7 mumol/l at the beginning and 4.6 +/- 1.8 mumol/l at the end). Strontium concentrations in blood plasma were respectively, in controls 0.22 +/- 0.06 mumol/l and in hemodialyzed patients 0.62 +/- 0.24 mumol/l before the dialysis session and 0.64 +/- 0.14 mumol/l at the end. The mean concentration of strontium in the dialysis fluid was the same before (0.49 +/- 0.11 mumol/l) and after the dialysis session (0.49 +/- 0.10 mumol/l), but a transfer between plasma and dialysis fluid was shown by individual changes. Some considerations about these results are put forward but their possible clinical consequences are not yet known.     

Selenium and glutathione peroxidase variations induced by polyunsaturated fatty acids oral supplementation in humans.

Serum and erythrocyte selenium, erythrocyte and platelet glutathione-peroxidase (GSH-Px) activities, and erythrocyte reduced glutathione (GSH) content were measured in 25 healthy adult individuals before and after daily supplementation with 20 ml of fish oil for 10 weeks. Serum-Se decreased from 0.83 +/- 0.01 mumol/l to 0.75 +/- 0.02 mumol/l (mean +/- S.E.M.) (P less than 0.01); erythrocyte-Se decreased from 4.39 +/- 0.17 nmol/g hemoglobin (Hb) to 2.83 +/- 0.15 nmol/g (P less than 0.001). GSH-Px activities increased both in erythrocytes (6.93 +/- 0.24 iu/g vs 8.18 +/- 0.27 iu/g Hb, P less than 0.01) and in platelets (69.2 +/- 2.8 iu/g vs 90.9 +/- 3.6 iu/g protein, P less than 0.001). The concentration of GSH in erythrocytes fell from 9.56 +/- 0.29 mumol/g Hb to 5.90 +/- 0.30 mumol/g Hb (P less than 0.001). The effects on plasma lipids were evident only for triglycerides (before 1.96 +/- 0.16 mmol/l, after 1.75 +/- 0.14 mmol/l, P less than 0.001). We hypothesise the enrichment of erythrocyte and platelet membranes with polyunsaturated fatty acids (PUFAs), following fish oil intake, can generate increased amounts of lipid peroxides and thus allosterically activate GSH-Px: with time this is harmful for the integrity of the enzyme molecule and Se release may result. We suggest that the Se status of individuals given PUFAs is assessed before and during intake; Se supplements should only be given when serum and/or erythrocyte Se are reduced.     

Cerebrospinal fluid levels of catechols in patients with neurogenic orthostatic hypotension

In multiple system atrophy (MSA) and pure autonomic failure (PAF), orthostatic hypotension (OH) results from deficient noradrenaline release from sympathetic nerves during standing. Post-mortem findings have indicated loss of central noradrenergic cells in both diseases. The present study sought in vivo neurochemical evidence for central noradrenergic deficiency in patients with OH due to MSA or PAF. A total of 28 patients with OH (18 with MSA; 10 with PAF) had cerebrospinal fluid and blood sampled for levels of noradrenaline and its neuronal metabolite dihydroxyphenylglycol. A control group of 44 subjects included 10 elderly normal volunteers, 10 patients with Alzheimer's disease, 18 patients with dysautonomia (postural tachycardia syndrome or neurocardiogenic syncope) and six patients with MSA in the absence of OH. Patients with OH had lower cerebrospinal fluid concentrations of noradrenaline (0.53+/-0.07 nmol/l) and dihydroxyphenylglycol (6.52+/-0.46 nmol/l) than did control subjects (0.90+/-0.09 and 9.64+/-0.46 nmol/l respectively; P =0.0001). The MSA+OH group had higher plasma levels of both catechols (noradrenaline, 1.31+/-0.16 nmol/l; dihydroxyphenylglycol, 5.08+/-0.43 nmol/l) than did the PAF group (noradrenaline, 0.38+/-0.08 nmol/l; dihydroxyphenylglycol, 2.53+/-0.30 nmol/l; P <0.001), despite similarly low cerebrospinal fluid levels. Among MSA patients, those with OH had lower cerebrospinal fluid levels of noradrenaline and dihydroxyphenylglycol than those without OH (noradrenaline, 1.71+/-0.64 nmol/l; dihydroxyphenylglycol, 10.41+/-1.77 nmol/l respectively; P =0.006). The findings are consistent with central noradrenergic deficiency in both MSA+OH and PAF. In MSA, central noradrenergic deficiency seems to relate specifically to OH.     

Relationships between androgen and estrogen sulfates in breast cyst fluid.

In 28 breast cyst fluids obtained from 20 patients (age 29-65 years) sodium, potassium and the sulfates (S) of estrone (E1), estradiol (E2), dehydroepiandrosterone (DHEA) and androsterone (A) were determined. The radioimmunoassays (RIA) used were validated for this particular biological fluid. According to electrolyte ratio (Na+/K+) the cyst fluids were subdivided into two groups: the first with low (less than 3) (n = 16) and the other with high (greater than 3) (n = 12) values. Markedly higher steroid sulfate levels were observed in the first group, the mean levels being: 147.7 nmol/l, 54.6 nmol/l, 108.1 mumol/l and 158.0 mumol/l for E1S, E2S, DHEAS and AS respectively. The mean levels in the second group were: 13.6 nmol/l, 6.7 nmol/l, 68.8 mumol/l and 33.6 mumol/l for E1S, E2S, DHEAS and AS, respectively. In the first group only E1S and E2S levels were significantly correlated (r = 0.51; P less than 0.05). Conversely, the steroid sulfate levels were significantly correlated with each other in the group with high electrolyte ratio. These data have confirmed preceding results and have clearly shown that breast cyst fluids with low electrolyte ratio contain more E2S than the other group. This finding might be correlated with the fact that patients with these breast cysts lined by with apocrine epithelium may be at a greater risk of breast cancer than those with the other type.     

Effect of aminophylline on plasma and urinary catecholamine levels during heavy leg exercise in healthy young men

1. Methylxanthines have been shown to elevate the basal plasma level and/or urinary excretion of noradrenaline (NA) and adrenaline (ADR) in healthy subjects. The present study addressed the hypothesis that the methylxanthine aminophylline also augments plasma and urinary catecholamines during increased sympathoadrenal activity. 2. Eleven healthy young men performed a maximal 2 h bicycle exercise twice, after double-blind intravenous administration of placebo or aminophylline. Femoral venous plasma and urinary concentrations of NA and ADR were analysed in samples representing basal state, exercise and recovery, using liquid chromatography with electrochemical detection. 3. Leg exercise induced eight- and six-fold increases in the plasma concentrations of NA and ADR, respectively, and seven- and four-fold increases in the urinary concentrations of NA and ADR, respectively, indicating that sympathoadrenal activity was considerably elevated. 4. After aminophylline (mean plasma concentration 20-35 mumol/l), the plasma concentrations of NA (P less than 0.001) and ADR (P less than 0.05) were independently higher at rest, during exercise and during recovery, in comparison to after placebo; the mean exercise plasma level of NA was increased by the drug from 13 +/- 1 to 21 +/- 2 nmol/l and the corresponding level of ADR from 2.1 +/- 0.4 to 2.9 +/- 0.5 nmol/l. Also urinary NA (P less than 0.01) and ADR (P less than 0.05) were elevated by aminophylline; the exercise concentrations of NA in the urine were 75 +/- 8 and 97 +/- 10 mumol/mol of creatinine after placebo and aminophylline, respectively, and the corresponding levels of ADR were 12 +/- 3 and 16 +/- 3 mumol/mol of creatinine, respectively.     

Determination of 3,4-dihydroxyphenylalanine (DOPA) in plasma and cerebrospinal fluid by high performance liquid chromatography with electrochemical detection

We report a reliable method for determining DOPA levels in plasma and cerebrospinal fluid. The method is based on complete conversion of DOPA to dopamine and quantification by HPLC-ECD of the dopamine formed. Lower limit of detection was 0.5 nmol/l. No differences in plasma DOPA levels were found between normal children (0-15 yr, n = 60), normal adults (n = 39) and patients with essential hypertension (n = 40) or Parkinson's disease (no DOPA therapy, n = 30). In normal individuals and in patients with essential hypertension venous plasma levels were higher than arterial levels (10.2 vs 9.3 nmol/l, p less than 0.001, V/A ratio 1.11 (SD 0.08), n = 15). Sympathetic stimuli (standing, tilting, bicycle exercise, tyramine) did not influence DOPA levels. In untreated depressed patients (n = 10) and in non-parkinsonian neurological patients (n = 12) cerebrospinal fluid levels of DOPA were 4.5 (SD 2.4) and 5.2 (SD 1.3) nmol/l respectively. A direct method for the measurement of DOPA by HPLC-ECD after deproteinization of plasma is also described and compared with the conversion method. Good agreement was found when plasma DOPA levels exceeded 0.25 mumol/l (y(conversion method) = 0.943x (direct method) + 0.118; n = 60; r = 0.985). The direct method, because of greater simplicity and the possibility of simultaneous measurement of the DOPA metabolite 3-O-methyldopa, is the method of choice with plasma samples from DOPA-treated patients. In non-DOPA treated individuals the conversion method is superior and has proved to be an accurate and sensitive method for the determination of DOPA levels in plasma and cerebrospinal fluid.     

Binding of corticotropin-releasing hormone (CRH) in maternal and fetal plasma and in amniotic fluid

Placenta secretes corticotropin-releasing hormone (CRH) into the maternal and fetal circulation, but a CRH binding protein in plasma may decrease its biological activity. Using a charcoal adsorption method we found that 92% of added 125I-Tyr-CRH was bound to a binding protein in the nonpregnant plasma, 72% in the plasma at term pregnancy, 90% in umbilical cord plasma, 82% in the amniotic fluid in the second and 25% in the third trimester. CRH added to plasma inhibited the binding of 125I-Tyr-CRH over the concentration range of 0.1-8.8 nmol/l in plasma and of 0.1-2.2 nmol/l in amniotic fluid. There was a significant negative correlation (R = -0.80) between the binding capacity of the CRH-binding protein and CRH concentration in maternal plasma. Plasma or amniotic fluid was incubated with 125I-Tyr-CRH and subjected to gel filtration on Sephadex G-50. The bound radioactivity was eluted at the region of Mr 25-40 kDa and the unbound radioactivity at the location of synthetic CRH. Bound and unbound CRH concentrations were determined using charcoal adsorption method and gel filtration on Sephadex G-50 in ten maternal plasma samples at the third trimester of pregnancy. Following mean percentages were found to be bound: charcoal method 61.9 +/- 6.80% (SE) and gel filtration 62.8 +/- 6.33%. We conclude that the bulk of CRH is bound to a binding protein in maternal and fetoplacental circulation, whereas at term pregnancy the role of the binding is small in amniotic fluid.     

Allantoin as a marker of oxidative stress in human erythrocytes

Abstract Background: Uric acid is the final product of purine metabolism in humans. It was determined that this compound has important antioxidative properties and it may be oxidized to allantoin by various reactive oxygen species. Therefore, the measurement of allantoin may be useful for the determination of oxidative stress in humans. Methods: We measured allantoin and uric acid in human plasma and erythrocytes obtained from patients with chronic renal failure before hemodialysis (n=30) and blood donors (n=30). We used a method based on selective isolation of allantoin from deproteinized plasma and erythrocyte lysate samples on AG 1-X8 resin and its derivatization to glyoxylate-2, 4-dinitrophenylhydrazone. Separation of glyoxylate-2, 4-dinitrophenylhydrazone from interfering substances was achieved on reversed phase HPLC with gradient elution and UV/VIS detection at 360 nm. Uric acid was determined by reversed phase HPLC with UV/VIS detection at 292 nm. Results: We found significant differences in allantoin and uric acid concentration between the patients with chronic renal failure and the control group both in plasma (20.5+/-6.5 mumol/L and 323.9+/-62.9 mumol/L vs. 2.1+/-1.1 mumol/L and 270.1+/-62.3 mumol/L, p<0.05) and erythrocytes [82.8+/-39.1 nmol/g hemoglobin (Hb) and 110.7+/-28.8 nmol/g Hb vs. 20.1+/-6.1 nmol/g Hb and 82.1+/-23.7 nmol/g Hb, p<0.05]. Conclusions: Significant higher levels of allantoin in both plasma and erythrocytes of patients with chronic renal failure indicate that allantoin may be used as a good marker of oxidative stress. Clin Chem Lab Med 2008;46:1270-4.     

Free choline in human plasma analysed by simple radio-enzymatic procedure: age distribution and effect of a meal

Ultrafiltration of plasma was shown to be a simple and rapid method to obtain a stable sample for direct measurement of free choline (Ch) in plasma by a radioenzymatic procedure. Free Ch was analysed in plasma from healthy volunteers fasted 12-15 h and 1 h after a meal. The free Ch concentration was found within narrow limits with a mean of 10.6 +/- 0.4 mumol/l in the fasted subjects and 11.5 +/- 0.3 mumol/l 1 h after a meal. The difference is significant (paired t test, P less than 0.01, n = 23). Dietary influence on the free Ch concentration in human plasma is suggested. In three newborn infants (1-3 min post partum) the Ch concentration in plasma from the umbilical vein was 24.5 +/- 1.9 mumol/l.     

Measurement of free choline concentrations in maternal and neonatal blood by micropyrolysis gas chromatography

The development of a micropyrolysis gas chromatographic technique for the determination of free choline in plasma and erythrocytes is reported as a means to clinically assess choline status. A micropyrolyzer syringe unit was fabricated and the method was standardized through repeated trials with a single plasma specimen. The interassay coefficient of variance was 1.3% for the standardized trials on plasma and erythrocytes. Choline status was assessed in control women, mothers and premature infants at birth and up to 14 days of life. At birth, plasma choline levels in the infants (32.2 +/- 5.5 mumol/l) were significantly higher than those in the mothers (12.9 +/- 2.6 mumol/l) and the control women (16.9 +/- 1.6 mumol/l). The infant's plasma choline concentrations had decreased significantly by 7 days of life and remained at the lower level at day 14, independent of the nutritional intervention administered.     

The effect of oral N-acetylcysteine on glutathione concentration in bronchoalveolar lavage of patients with fibrosing lung diseases

Glutathione (GSH) is an important constituent in protecting the cellular elements within the lower respiratory tract against oxydants. We measured GSH in bronchoalveolar lavage fluid (BAL) in twelve patients with lung fibrosis and compared the data to eight healthy controls. GSH in BAL was 0.71 +/- 0.34 mumol/l in patients with lung fibrosis and 0.88 +/- 0.35 mumol/l in the controls (p greater than 0.05). After relating GSH on the volume of the epithelial lining fluid (ELF), determined by the urea method, GSH/ELF was 93 +/- 71 mumol/l in lung fibrosis and significantly different (p less than 0.005) from 387 +/- 240 mumol/l in the controls. In seven patients (four lung fibrosis, one desquamative pneumonitis, two asbestosis) GSH in BAL was determined before and after seven days of medication with 1800 mg N-acetylcysteine/day. We observed a significant increase of GSH in BAL from 0.93 +/- 0.46 to 1.56 +/- 0.92 mumol/l. Our observations confirm that in patients with lung fibrosis the protective ability against oxydants is diminished and that one parameter of the antioxydative capacity (GSH) can be increased in BAL by oral administration of N-acetylcysteine. Further studies are necessary to investigate the clinical and therapeutic implications of our findings.     

Determination of thiocyanate in plasma and saliva without deproteinisation and its validation as a smoking parameter

A simple photometric procedure was developed for the determination of thiocyanate (SCN-) in plasma and saliva without deproteinisation or dialysis. Fe(III) ions form a red coloured complex with SCN- with a maximum absorbance at 460 nm. Mercury(II) nitrate is used to run a sample blank. A manual and an automated version (COBAS BIO) of the method is described. The method is linear up to 5000 mumol/l SCN-. The CV of the between-run precision is 2.8-8% for the manual and 2.6-6.6% for the automated method. The SCN- plasma concentration was 21-134 mumol/l in nonsmokers and 44-260 mumol/l in smokers. In mixed saliva, the concentration is much higher than in plasma: 1.57-5.5 mumol/l in smokers and 0.79-3.9 mumol/l in nonsmokers. Plasma, but not oral fluid SCN-, is a valuable parameter for studying smoking habits in population surveys. However, its use is limited. In our experience only heavy smokers can be distinguished from nonsmokers.     

Plasma glycolate concentrations under conditions of dialysis

Plasma glycollate and oxalate concentrations were measured in 20 patients undergoing chronic haemodialysis treatment. The mean plasma glycollate level was 173.7 +/- 52.9 mumol/l, which was not significantly different from the normal value (means = 145.8 +/- 37.8 mumol/l). The mean plasma oxalate concentration (means = 128.7 +/- 25.6 mumol/l) was about 8 times higher than the value found in normal volunteers (means = 16.8 +/- 6.0 mumol/l). During haemodialysis lasting for 6 hours the plasma oxalate concentration decreased by 53.5%. However, no decline in plasma glycollate levels was noted. Since glycollate was not found in ultrafiltrates obtained in vivo, it is concluded that glycollate is not eliminated during haemodialysis treatment.     

Sex differences in the hepatic uptake of sulphobromophthalein in the rat

Sex difference in the hepatic uptake of sulphobromophthalein (BSP) was investigated in male and female rats in three different experimental models. In the intact animal the BSP plasma disappearance rate was significantly higher (P less than 0.01) in females than in males when 0.15 or 1.5 mumol/kg body wt. was injected. Comparable values were found at the highest BSP dose (15 mumol/kg body wt.) used. In the perfused liver, the first-pass hepatic extraction and the uptake velocity were significantly higher (P less than 0.001) in female rats at low BSP doses (0.3-750 mumol/g of liver) whereas identical values were found at higher concentrations. In hepatocytes isolated by collagenase perfusion, the BSP uptake occurs via two different uptake sites in both sexes. The Km of the high affinity sites was lower in females than in males (3.67 +/- 0.58 vs 7.24 +/- 0.68 mumol/l, P less than 0.001) whereas Vmax. showed comparable values (2.70 +/- 0.36 vs 2.47 +/- 0.45 nmol of BSP/mg of protein, NS). In contrast, no difference was found in the kinetic parameters of the low affinity sites (Km 50.6 +/- 31.1 vs 61.0 +/- 17.5 mumol/l; Vmax. 21.9 +/- 13.2 vs 25.0 +/- 3.6 nmol of BSP/mg of protein, mean +/- SD, NS, females and males respectively). Taken together these data show that low doses of BSP are taken up by the liver more efficiently in female than in male rats and are consistent with a sex-related difference in the affinity but not in the number of the BSP high affinity uptake sites.     

Kawasaki disease and oxidative metabolism

The plasmid lipid peroxidation products: malonyldialdehyde (MDA), organic hydroperoxides (OHP) and zinc peripheral values were analysed in seven children with Kawasaki disease, in the acute phase, before and after treatment with intravenous immunoglobulins. In the acute phase, plasma levels of MDA (2.95 +/- 0.30 mumol/l, control group: 2.52 +/- 0.08 mumol/l) and OHP (235 +/- 65 mumol/l, control group: 120 +/- 10) were increased (p less than 0.05 and p less than 0.01). Moreover, plasma zinc levels were significantly decreased (10 +/- 2.15 mumol/l versus 15.6 +/- 2.5 mumol/l in control group, p less than 0.05). After treatment with immunoglobulins, MDA, OHP and plasma zinc levels returned to normal. The parallel normalisation of clinical injury draws attention to the possible role of oxygen intermediates in connective tissue degradation and in the pathogenesis of vascular abnormalities in Kawasaki disease.     

A new method for the determination of L-dopa and 3-O-methyldopa in plasma and cerebrospinal fluid using gas chromatography and electron capture negative ion mass spectrometry

L-3-(3,4-Dihydroxyphenyl)alanine (DOPA) and its 3-O-methyl metabolite (OMD) were measured in plasma and cerebrospinal fluid by a new assay which combines N,O-acetylation of amino acids in aqueous media, preparation of pentafluorobenzyl esters under anhydrous conditions, and analysis by gas chromatography-electron capture negative ion mass spectrometry. The N,O-acetyl, carboxy-PFB derivatives gave abundant carboxylate anions ([M-CH2C6F5]-) which were suitable for sensitive analysis using selected ion monitoring. Plasma and CSF samples were sufficiently purified by a simple organic solvent extraction. Analytical recovery for DOPA was 100.2 +/- 3.7% at the level of 100 nmol/l. Analysis of DOPA in plasma was performed with a relative standard deviation of 5%. The limit of quantitation in plasma and CSF was at the sub-nmol/l level. In healthy adults, DOPA concentration in plasma was 9.0 +/- 2 nmol/l (n = 11) and in CSF 3.5 +/- 0.9 nmol/l (n = 9). The concentration of OMD in plasma was 99.1 nmol/l (pool of 24 samples) and 15.3 nmol/l in CSF (pool of 12 samples). Measurement of 5-[2H]DOPA and 5-[2H]OMD in plasma of a healthy individual who had been orally loaded with 3,5-[2H2]tyrosine (150 mg kg body wt) was possible for several hours after the load.     

Accurate and specific HPLC assay of coproporphyrinogen III oxidase activity in human peripheral leucocytes

An HPLC method is described for the assay of coproporphyrinogen III oxidase in human leucocytes. The optimal pH for the assay was 6.5-7.0 and the Km for coproporphyrinogen III was 0.12 +/- 0.021 mumol/l. The mean activity in 28 apparently health subjects was 0.249 (2 SD range 0.130-0.368) nmol/h per mg protein. In two patients with hereditary coproporphyria, the activities were 0.029 and 0.078 nmol/h per mg protein.     

The effects of cardiopulmonary bypass on androgen hormones in coronary artery bypass surgery

The effects of testosterone on coronary vasomotor regulation have been described by several recent reports. Here we investigated changes in serum androgen levels during and after cardiopulmonary bypass (CPB) in patients who had undergone coronary artery bypass surgery. Serum luteinizing hormone, free testosterone and dihydroepiandrestenedione sulphate (DHEA sulphate) levels were evaluated in 38 male coronary artery bypass surgery patients using a chemical immunoassay technique. All hormone levels were corrected to account for haemodilution. Serum-free testosterone level decreased significantly during weaning from CPB (from 15.7 +/- 4.2 nmol/l to 6.2 +/- 2.8 nmol/l), and an even greater decrease was observed in the first post-operative day (5.4 +/- 3.1 nmol/l). On the seventh post-operative day, free testosterone levels reached a normal value (11.8 +/- 5.5 nmol/l), although they were still significantly lower compared with the pre-operative value. There were slight alterations in serum DHEA sulphate levels, although the only significant decrease occurred from the first to the seventh day post-operation (from 4.7 +/- 2.2 mumol/l to 3.7 +/- 1.8 mumol/l, respectively). Serum luteinizing hormone levels were decreased during weaning from CPB (from 4.8 +/- 2.1 mIU/ml to 3.9 +/- 1.8 mIU/ml), but increased rapidly to the pre-operative value (5.5 +/- 2.5 mIU/ml) at the first post-operative day. These results show that CPB affects serum luteinizing hormone, free testosterone and dihydroepiandrestenedione sulphate levels. The free testosterone level decreases significantly both during and after CPB surgery.