Quantification of liver iron concentration with magnetic resonance imaging by combining T1-, T2-weighted spin echo sequences and a gradient echo sequence.

BACKGROUND: The aim of the study was to quantify hepatic iron by MRI for practical use. METHODS: In twenty-three patients with various degrees of iron overload, measurements were carried out with a 1.5 Tesla MR unit. A combination of pulse sequences (T1, T2 and gradient echo) enabled us to quantify smaller amounts of liver iron as accurately as larger amounts of liver iron. The gradient echo sequence provided us with a good correlation when detecting smaller amounts of iron in the liver where the T1 sequence provided a good correlation when larger amounts of iron were present. RESULTS: The combination of the three sequences showed a nice correlation (r=-0. 93, P<0.001) and provided us with an accurate estimate of the liver iron content (LIC). This correlation was achieved with a LIC from the lower range of normal up to LIC of 146 mmol/kg dry weight, which seems the highest measurable liver iron content for a 1.5 Tesla MRI. Measuring in the lower range makes it possible to decide whether further invasive diagnostic investigations by a liver biopsy are indicated. CONCLUSION: MRI is a useful tool to quantify iron overload non-invasively. In cases where a liver biopsy is hazardous MRI can easily be used to obtain reliable, quantitative information about the initial LIC. Quantification by MRI could also be used for follow up of the iron content during depletion treatment by phlebotomy or iron chelation. The stronger the magnet the more sensitive the detection of concentrations up to 150 mmol/kg is. A semi-quantitative judgement will only be possible with severe iron overload over 150 mmol/kg. If such an iron excess is found, a liver biopsy should be performed to exclude cirrhosis.

Major clinical events, signs and severity assessment scores related to actual survival in patients who died from primary biliary cirrhosis. A long-term historical cohort study.

BACKGROUND/AIMS: One of the prognostic methods for survival in primary biliary cirrhosis (PBC) is the Mayo model, with a time-scale limited to 7 years. The aim of our study was to assess how major clinical events, signs, several severity assessment methods and Mayo survival probabilities fit in with actual patient survival, by using yearly observations until 0.5 years before patient death from PBC. METHODOLOGY: Data of 32 patients dying from PBC were collected prior to death at -0.5, -1, -2 etc. years (median: -5 years, range: -16 to -0.5 years). Major events registered were: first occurrence of ascites, upper gastrointestinal bleeding or manifest hepatic encephalopathy and signs, first observation of spider naevi or purpura. Severity assessment methods applied (all with scores and classes) were: Mayo (M), Child-Campbell (C), Pugh-Child (P), Pugh-Child-PBC (PP), 'Child-Pugh' (CP), and Ascites Nutritional State-Child (ANS). Fifty percent survival estimates were calculated from Mayo scores. Severity assessment method variables were: ascites (C, P, PP, CP, ANS), encephalopathy (C, P, PP, CP), nutritional state (C, ANS), edema (M), age (M), serum albumin (M, C, P, PP, CP), bilirubin (C, M, P, PP, CP), and prothrombin time (M, P, PP, CP). RESULTS: In 27 out of 32 patients a major event occurred, always between -6 and -0.5 years (median: -1 year) and, never between -16 and -7 years (p < 0.0001). A sign was first observed in 30/32 between -14 and -0.5 years (median: -2 years). Compared to the total population, a sign, and even more so, an event indicated a shorter survival (p = 0.004 and p = 0.0002, respectively). The median 50% estimated survival (predicted by the Mayo model) fitted the actual survival from -6 to -0.5 years (r = -0.7, p < 0.0001), but not from -16 to -7 years (r = -0.1, p = 0.4). All -6 to -0.5-year severity scores correlated (p < 0.0001) both with actual survival (M, C, P, PP, and CP r = 0.7; ANS r = 0.5) and with estimated M 50% survival (C, P, PP, CP r = -0.9; ANS r = -0.6; M score: -0.99), but none with actual survival from -16 to -7 years, except for M, slightly (r = -0.3, p = 0.04). A nomogram for mean C, CP, M and ANS scores related to actual survival was constructed for the -6 to -0.5-year period. The C and CP classes A, B, and C did not appear to distinguish sufficiently into actual survival, whereas the M classes did. CONCLUSIONS: The occurrence of a major event appeared to exclude survival over 6 years. In these final 6 years, Child-Campbell, Mayo and Pugh scores correlated equally well with actual survival and better than Ascites/Nutritional State score. In our PBC patients, Campbell was an excellent alternative for Pugh; for Pugh, the original Child-Turcotte variable limits were fully sufficient.

Respiratory ammonia output and blood ammonia concentration during incremental exercise.

The aim of this study was to investigate whether the increase of ammonia concentration and lactate concentration in blood was accompanied by an increased expiration of ammonia during graded exercise. Eleven healthy subjects performed an incremental cycle ergometer test. Blood ammonia, blood lactate and the amount of expired ammonia were measured until 30 minutes post exercise. The expired air was guided through a flow chamber filled with a sulphuric acid solution to trap the expired ammonia. Blood ammonia, blood lactate increased more than proportionally and the amount of expired ammonia (in micromol/min) increased exponentially with the workload. Post-exercise the amount of expired ammonia decreased within a few minutes back to pre-exercise levels while the concentrations of lactate and ammonia in blood decreased much more slowly and were still elevated after 30 minutes of recovery. We conclude that the more than proportional increase of ammonia and lactate during graded exercise, is accompanied with an exponential increase of expired ammonia output. Faster and more accurate ammonia gas detection techniques are necessary to quantify more precisely the respiratory ammonia output during graded exercise.

Helicobacter pylori and ammonia concentrations of whole, parotid and submandibular/sublingual saliva.

The aim of this study was to determine the ammonia concentration in whole, parotid and submandibular/sublingual saliva of healthy volunteers using the indophenol direct method. It also investigated the hypothesis that higher saliva ammonia concentrations are associated with the presence of Helicobacter pylori (H. pylori) in the oral cavity. In healthy volunteers, the mean ammonia concentration of whole saliva (2574 mumol/l) was significantly higher (P < 0.0001) than the mean ammonia concentration of both parotid (238 mumol/l) and submandibular/sublingual (355 mumol/l) saliva. In whole saliva, no difference in ammonia concentration was found between healthy controls and dyspeptic patients (mean ammonia values 2574 and 2489 mumol/l respectively, P = 0.7). In addition, no significant differences were observed in the salivary ammonia concentration between dyspeptic patients with and without H. pylori carriage. It is concluded that the ammonia concentration in parotid and submandibular/sublingual saliva does not differ, but is significantly lower than the ammonia concentration of whole saliva. This difference is not due to carriage of H. pylori with its strong urease activity. Therefore, the determination of ammonia in whole saliva is an inappropriate screening test for patients being at risk for (chronic) gastritis and peptic ulcer disease.

Determination of ammonia in cerebrospinal fluid using the indophenol direct method.

We have determined ammonia in cerebrospinal fluid (CSF) with the indophenol direct method. The results were compared with an enzymatic method. The method is very simple, and precision (coefficient of variation 1.6%) and linearity (r = 0.9999, p < 0.001) of the method are excellent. The recoveries of the method are very good (within-sample recovery: range 88-93, median 93%; between-sample recovery: 88-93, median 91%). In a population of 23 neurological patients not suffering from liver disease, the reference values ranged from 8 to 26, median 18 microM. Males and females did not differ (p = 0.5). The values obtained with the indophenol method were equal to the enzymatic method (range 9-28, median 18 microM, p = 0.6). On storage in the deep freeze (-20 degrees C), there was no change in CSF ammonia concentration for at least 1 mo. When stored at 4 degrees C (refrigerator), ammonia determinations have to be performed within 2 d. CSF storage at room temperature results in artificially elevated ammonia levels and should be avoided.

Primary biliary cirrhosis: Dutch application of the Mayo Model before and after orthotopic liver transplantation.

BACKGROUND/AIMS: A retrospective study of primary biliary cirrhosis (PBC) was performed to study the Original Mayo Model for predicting survival by a Dutch data-set of patients, presentation of disease progression; assessment of liver transplantation, prediction of post-transplantation survival; and the addition of two laboratory variables to the Original Mayo Model. MATERIALS AND METHODS: Survival of 83 patients, 37 of whom underwent transplantation, were studied. Mean follow-up was 6.0 +/- 0.45 SEM years. Risk score at diagnosis, platelet count, and serum sodium were analyzed in a Cox model. RESULTS: The Original Mayo Model estimated survival for low-, medium-, and high-risk groups accurately and it also presented disease progression. Baseline Mayo risk score in a Cox model had a regression coefficient of 1.01, indicating an excellent predictor p < 0.0001. Platelet count was a predictor of survival (p < 0.002), whereas serum sodium did not (p = 0.67). A new model combined of the Original Mayo risk score and platelet count predicted survival in high-risk patients somewhat better compared to the Original Mayo Model. With both models, liver transplantation had a significant beneficial effect on survival (p < 0.001). The scores revealed no significant influence (p = 0.47) for overall post-transplantation survival. CONCLUSIONS: The Original Mayo Model remains the model of choice for patients with PBC for prognostication from 3-8 years, is a useful tool in the assessment of liver transplantation but not an indicator of post-transplantation survival. Platelet count showed to have additional prognostic value. A new model combined of platelet count and the Original Mayo risk score did predict survival in high-risk groups slightly better compared to the Original Mayo Model.

High prevalence of hepatitis G virus after liver transplantation without apparent influence on long-term graft function.

BACKGROUND/AIMS: Hepatitis G virus is a recently characterized transfusion-transmissible RNA virus. Its pathogenicity remains to be established. We studied its prevalence in liver transplant patients and assessed the long-term influence on the liver graft. METHODS: Thirty-nine adult patients without hepatitis B or C were included; median follow-up was 8 years (range 1-17). Serum samples from before and late after transplantation were investigated for the presence of HGV-RNA. HGV-RNA was detected by cDNA-PCR, using primers from the NS3 region of the viral genome. The latest available yearly liver biopsy was assessed in a coded fashion according to established histological criteria. The outcome in the HGV-positive patients was compared with the outcome in the HGV-negative patients with respect to liver tests and liver histology. RESULTS: The prevalence before and after transplantation was 15.4 and 43.6%, respectively. Liver test results and liver histology did not differ between the HGV and non-HGV groups. In both groups more than 50% of the patients showed normal histology. Mild portal and/or lobular inflammation tended to be more prevalent in the non-HGV group (no statistical difference). CONCLUSIONS: HGV infection is highly prevalent in liver transplant patients. In the absence of co-infection with hepatitis B or C virus, no long-term negative influence on the graft occurs.

Whole blood and plasma water in health and disease: longitudinal and transverse observations and correlations with several different hematological and clinicochemical parameters.

In a healthy reference population, hemoglobin (Hgb) and hematocrit (Hct) have been proposed as surrogate markers for whole blood water (WBW). We have extended this study under different physiological and pathological conditions in two longitudinal series, viz. (1) acute hyper- and hypohydration experiments in a healthy individual and (2) three athletes running 5 km each, and in three transverse series, viz. (3) a young reference population (n = 97, 49 females), (4) an old reference population (n = 37, nine females) consisting of inhabitants of a nursing home and (5) cardiac, hematological and renal patients including severe anaemia, polycythaemia and abnormal protein levels (n = 50, 25 females) with suspected hydration disturbances. The only sex difference found was a lower WBW in males in the young reference group. The percentage change of PW was less than that of WBW. In all five groups together (n = 293) WBW correlated closely (P < 0.0001) with Hgb and Hct (both r = -0.95) and with erythrocyte count (r = -0.85), whereas PW correlated with total protein (Tprot) (r = -0.84). In the longitudinally studied groups (1) and (2) WBW also correlated (P < 0.0001) with cholesterol, Ca, Tprot, albumin, platelets, globulin and white blood cells (r +/- 0.98-0.37), while PW correlated (P < 0.0001) not only with the same clinicochemical parameters but also with Hct, Hgb and red blood cells (r +/- 0.98-0.44). The homeostasis of PW is more narrowly regulated than that of WBW. Hgb, Hct and erythrocyte count reflect WBW and Tprot reflects PW also under disease conditions. WBW (mass%) can be calculated from Hgb and Hct using the formulae: -0.09 x Hgb (g/l) + 91.7 and -28.6 x Hct (v/v) + 91.8 and PW (mass%) from Tprot using the formula: -0.09 x Tprot (g/l) + 97.6. Other correlations were observed only in a longitudinal setting and presumably are due to concentration and dilution.

Lactate and ammonia concentration in blood and sweat during incremental cycle ergometer exercise.

It is known that the concentrations of ammonia and lactate in blood increase during incremental exercise. Sweat also contains lactate and ammonia. The aim of the present study was to investigate the physiological response of lactate and ammonia in plasma and sweat during a stepwise incremental cycle ergometer exercise test in ten subjects. During this test lactate and ammonia were measured in blood obtained from the earlobe and in sweat collected in a bag attached to the back of the subject. At the end of each interval this bag was emptied for measuring lactate and ammonia. A disproportional increase in the concentration of lactate and ammonia in blood was found, in sweat a disproportional decrease. The lactate concentrations in sweat were higher than those in blood. We hypothesise that lactate in sweat is produced from glycogen granules of the clear cell of the eccrine gland. This lactate production results in acidification of sweat, which facilitates the diffusion of ammonia from eccrine duct cell to duct lumen. It is uncertain how far duct cell ammonia originates from plasma, the duct cell itself might produce ammonia. Part of the ammonia in sweat could come from the breakdown of urea by skin bacteria.

Arterial ammonia with Blood Ammonia Checker II and with indophenol reaction to assess presence of hepatic encephalopathy.

Hepatic encephalopathy (HE) is associated with elevated arterial ammonia levels. The relationship is variable, in part due to ammonia methodology. One method, based on the indophenol reaction (IPh), is interfered with a number of amino acids including all aromatic amino acids. We have determined arterial ammonia simultaneously with the Blood Ammonia Checker II (BAC) as reference method and with the IPh method. The difference BAC-IPh, mumol/l, was assumed to express the interference in the indophenol method (IFI) by amino acids. It may be positive or negative. The aim was to establish the value of BAC in comparison with IPh in the diagnosis of liver disease and overt HE and to assess any added value of IFI. Of two reference groups without disturbances, A (n = 39) had not and B (n = 13) had encephalopathy. Group C consisted of 125 liver patients (34 no cirrhosis, 91 cirrhosis) of which 55 had no manifest HE (C:HE-) and 70 had HE (C:HE+). Median BAC ammonia nitrogen (NH3-N), mumol/l: A 21, B 35, C 80, C:HE - 57 and C:HE+ 98 (A < B < C and A < B < C:HE - < C:HE +, P < 0.001). Median IPh NH3-N, mumol/l: A 27, B 30, C 30, C:HE - 25 and C:HE + 35 mumol/l (A = B = C and C:HE - < C:HE+, P < 0.01). IFI medians: A -6, B 3, C 40, C:HE - 29 and C:HE + 58 mumol/l (A < B (P < 0.05) < C (P < 0.0001); A, B < C:HE - and C:HE+; C:HE- < C:HE + (all P < 0.0001)). While BAC correlated weakly with IPh in the (sub)groups C, C:HE-, C:HE+ (r = 0.3, 0.3, 0.4, P < 0.05), it correlated strongly with IFI (r = 0.9, 0.9, 0.8, P < 0.0001). There was no correlation between IPh and IFI. BAC, as well as IFI, could discriminate all liver patients (C) from both reference groups A and B with 100% positive likelihoods. BAC, IPh and IFI could discriminate between HE- and HE+. To differentiate cirrhosis from non-cirrhosis the specificity of IPh was uniformly high and the sensitivity satisfactory, whereas BAC had a high sensitivity but an insufficient specificity. In conclusion, in blood, BAC is the ammonia determination of choice. It differentiates between reference groups (encephalopathic or not) and liver disease and the more so HE. The combination of BAC and IPh (indicating IFI) may eventually be shown useful to rapidly assess the severity of underlying liver disease in HE patients. In other biological fluids, IPh is excellent when the inhibiting influence of non-protein nitrogen substances is absent or can be eliminated.

Determination of ammonia in ear-lobe capillary blood is an alternative to arterial blood ammonia.

Blood ammonia determination is a laboratory test to diagnose hepatic encephalopathy. Arterial blood is superior to peripheral venous blood ammonia because of ammonia metabolism in muscle. We have compared capillary with arterial whole blood ammonia as capillary sampling is an attractive alternative. Ear-lobe capillary blood ammonia (ECA) was determined in all 173 persons studied, fingertip capillary blood ammonia (FCA) in 46 of these and arterial blood ammonia (AA) in 113. Of the 173, 60 were healthy (H), 64 were patients, not liver diseased (NLD) and 49 had liver disease (LD). Reference values, median and ranges, mumol NH3-N/l: AA, NLD, n = 64: 17 (7-42); ECA, H = NLD (P = 0.9), n = 124: 20 (7-45); FCA, H = NLD (P = 0.8), n = 33: 70 (29-151). Within the NLD group (n = 64) AA values (range 7-42) were little but significantly lower than the ECA values (range 7-45, P = 0.002). FCA NLD > AA NLD (n = 14, P < 0.0001); FCA H+NLD > ECA (n = 33, P < 0.0001). AA correlated very well with ECA, r = 0.87 (n = 113, P < 0.0001) and less well with FCA, r = 0.56 (n = 27, P < 0.01). ECA correlated with FCA, r = 0.51 (n = 46, P < 0.001). Ear-lobe capillary blood ammonia thus accurately reflects arterial ammonia and is an attractive alternative. The higher fingertip ammonia may be due to contamination with ammonia-rich sweat from finger grooves, regardless of the precautions taken.

Evaluation and clinical application of the Enterotest for the determination of human biliary porphyrin composition.

The Enterotest string test is an easy and non-invasive method for sampling duodenal fluid, which has been successfully used for the analysis of duodenal microflora, as well as biliary bile acid and lipid composition. The method was evaluated for determination of porphyrins in duodenal bile in normal subjects and subjects with porphyria, following cholecystokinin induced gall bladder contraction; it is known that analysis of biliary porphyrins is more discriminatory for the diagnosis of asymptomatic porphyria than their analysis in faeces or urine. Moreover, serial analysis of bile from patients with erythropoietic protoporphyria may help in establishing their ability to secrete protoporphyrin in bile and to assess effects of treatment. The binding of various porphyrins to Enterotest strings was investigated by incubating pieces of the string in different human bile samples with low to very high porphyrin concentrations, followed by HPLC analysis of porphyrins both in the native bile and in extracts obtained from the strings. No differences between porphyrin composition in native bile and extracts were observed. Duodenal fluid obtained by means of the Enterotest from volunteers not receiving cholecystokinin showed large variations in porphyrin patterns not resembling those of native bile. Mesoporphyrin, a secondary porphyrin derived from protoporphyrin by bacteria, was often detectable. These data indicate that the duodenal content without cholecystokinin injection does not reflect biliary porphyrin composition. The presence of mesoporphyrin in the whole intestinal tract, but not in serum and bile, suggests that there is no enterohepatic circulation of secondary porphyrins. There was close agreement between the porphyrin ratios found with the standard duodenal intubation technique and the Enterotest, performed simultaneously in one healthy volunteer after induction of gall bladder contraction by cholecystokinin. From these experiments, it was concluded that fluid adsorbed to the Enterotest string after gall-bladder contraction can be used to determine biliary porphyrin composition. Since duodenal bile is diluted gall bladder bile, variable porphyrin concentrations were found when applying the Enterotest in combination with cholecystokinin in the same subject on successive days. However, porphyrin ratios, such as the protoporphyrin to coproporphyrin I ratio, were relatively constant. In subjects with symptomatic variegate porphyria, the Enterotest showed highly aberrant porphyrin patterns, with increased protoporphyrin to coproporphyrin I ratios and, in addition, the presence of some unknown porphyrins. A deviating biliary protoporphyrin/coproporphyrin I ratio in one patient appeared to be a useful diagnostic index for the presence of latent variegate porphyria (or variegate porphyria in remission).(ABSTRACT TRUNCATED AT 400 WORDS)

Nitrite and nitrate determinations in plasma: a critical evaluation.

Plasma nitrite and nitrate determinations are increasingly being used in clinical chemistry as markers for the activity of nitric oxide synthase and the production of nitric oxide radicals. However, a systematic evaluation of the determination of nitrite and nitrate in plasma has not been performed. In this study the recovery and stability of nitrite and nitrate in whole blood and in plasma, the relation between nitrite and nitrate concentrations in plasma, and possible sources of artifacts were investigated. The main conclusions are: (a) Recovery of nitrite and nitrate from plasma is near-quantitative (87%) and reproducible; (b) nitrite and nitrate are stable in (frozen) plasma for at least 1 year; (c) nitrite in whole blood is very rapidly (> 95% in 1 h) oxidized to nitrate, and therefore plasma nitrite determination alone is meaningless; (d) the ranges of nitrite and nitrate concentrations in plasma samples of 26 healthy persons are 1.3-13 mumol/L (mean 4.2 mumol/L) and 4.0-45.3 mumol/L (mean 19.7 mumol/L), respectively; (e) plasma nitrite and nitrate concentrations were not correlated (nitrite as % of total nitrite + nitrate varied from 3.9% to 88% in plasma samples); and (f) plasma samples should be deproteinized, and background controls for each sample should be included in the assay, to avoid measuring artifactually high nitrite and nitrate concentrations in plasma.

Determination of porphyrins in bile using high performance liquid chromatography and some clinical applications.

A simple and fast HPLC method for the determination of porphyrins in bile without extraction is described. Porphyrins were determined in bile from control subjects and from patients after orthotopic liver transplantation, including three patients with erythropoietic protoporphyria. It was found that: 1) coproporphyrin I is the predominant porphyrin in bile of controls, accompanied by some coproporphyrin III and protoporphyrin, whereas protoporphyrin mostly but not always is the predominant porphyrin in the bile of erythropoietic protoporphyria patients. In two of the three erythropoietic protoporphyria patients, the bile contained a hundred times more protoporphyrin than that of non-porphyric orthotopic liver transplantation patients. The third erythropoietic protoporphyria patient remained cholestatic and was unable to excrete sufficient amounts of protoporphyrin. 2) All investigated bile samples contained no secondary porphyrins derived from protoporphyrin, i.e. no deutero-, pempto-, or mesoporphyrin. Even when extracts of bile and serum were concentrated fifty to a hundred times, no traces of deutero-, pempto- and mesoporphyrin were detected. This complete absence of secondary porphyrins suggests that an enterohepatic circulation of dicarboxylic porphyrins from the distal gastrointestinal tract does not exist. 3) The HPLC chromatograms contain peaks from unknown compounds. No correlation between porphyrins and these compounds was found. Porphyrin profiles were followed in the bile of some orthotopic liver transplantation patients. Three episodes are recognizable. During the first three days after orthotopic liver transplantation there is a very high coproporphyrin excretion. There is then a lag of one to three weeks, in which no or very low porphyrin concentrations are detectable, followed by the restoration of normal biliary porphyrin patterns.

Antimitochondrial antibody profiles in patients with primary biliary cirrhosis before orthotopic liver transplantation and titres of antimitochondrial antibody-subtypes after transplantation.

Four antimitochondrial antibody profiles (A-D) have been defined in primary biliary cirrhosis according to the presence of antibodies to M2, M4, M8, and M9 in ELISA and the complement fixation test: A: anti-M9 positive in ELISA and western blot, B: anti-M9 and/or anti-M2 positive in ELISA, C: anti-M2, -M4 and/or -M8 positive in ELISA, D: anti-M2, -M4, and/or -M8 positive in ELISA and complement fixation test. These profiles predict the outcome of primary biliary cirrhosis in the early stages and reflect differences in the natural course of the disease (benign versus progressive). In this study sera from 29 patients with advanced primary biliary cirrhosis who had received liver transplant were retested before and after orthotopic liver transplantation. Twenty-eight were antimitochondrial antibody/anti-M2 positive, and one patient had only antibodies to nuclear dots in the immunofluorescence test on cell cultures. When the antimitochondrial antibody-profiles in these 28 anti-M2 positive patients were analysed, it became evident that 26 of them belonged to subgroup C or D before orthotopic liver transplantation. Two patients had profile B; one had high titres of antinuclear and smooth muscle antibodies indicating an overlap syndrome between primary biliary cirrhosis and autoimmune chronic active hepatitis. The other patient had antibodies to nuclear dots in association with anti-M2. None of the patients had profile A. Antibody titres were studied after orthotopic liver transplantation in 23 of the 28 patients who survived for 1 to 13 years.(ABSTRACT TRUNCATED AT 250 WORDS)

Gravimetric determination of the water concentration in whole blood, plasma and erythrocytes and correlations with hematological and clinicochemical parameters.

We have assessed gravimetric methods for determination of intravascular water, established whole blood-, plasma- and erythrocyte water reference values in a healthy volunteer group (n = 97, 48 females) and correlated these variables with 30 simultaneous hematological, clinicochemical and body parameters. The water standard was 55.56 mol/kg = 100 mass %. For erythrocyte water determination three methods were evaluated: 2 indirect methods were easy to perform, the third, using a hematocrit centrifuge, was the most reliable. Imprecision (within-batch coefficient of variation (CV), %) was excellent: whole blood 0.2, plasma 0.1, erythrocytes 0.7-2.2 and recoveries (means, %) 99.7-100.1. Serum water was found to be slightly higher than plasma water. Volunteer group, mean reference values, mass %: whole blood water 79.7, plasma water 91.2, erythrocyte water, three methods 66.2, 64.6 and 64.2, respectively. Females had mean 1.6 mass % higher whole blood water and 0.9-1.0 mass % higher erythrocyte water than males with no difference in plasma water. In the volunteer group whole blood water correlated strongly with hematocrit (r = -0.96), hemoglobin (r = -0.94) and erythrocytes (r = -0.85) and centrifuge hematocrit (r = -0.91). Plasma water correlated strongly with plasma total protein (r = -0.74, all correlations P < 0.001). Hemoglobin and hematocrit can serve as surrogate parameters for whole blood water when water determination is not available; total protein reflects plasma water.