ACTA ENDOCRINOLOGICA (BUC)

Introduction: The necessity to determine the levels of pituitary hormones in the cerebrospinal fluid is motivated both by difficulties in the diagnosis of different neuroendocrine disorders, as well as by research purposes such as understanding the transport mechanisms of hormones through the blood brain barrier.\r\nObjective: The aim of this study was to compare the results obtained with different immunometric methods for some pituitary hormones in the serum and CSF in patients with neuroendocrine tumors.\r\nMaterials and methods: The levels of LH, FSH, PRL and TSH were determined simultaneously in the serum and CSF with 3 different immunometric methods using commercial kits: IRMA, FIA and Chemiluminescence for 36 patients. The validation of IRMA method on CSF samples was performed using the dilution test.\r\nResults: The dilution test - as one of validation criteria of an immunoassay - proved that the results obtained in the CSF using the commercially available kits were correct. This enables the use of this sensitive method to accurately demonstrate abnormal levels of pituitary hormones beyond the blood-brain barrier. The correlation between IRMA and FIA: Hormonal concentrations values obtained by IRMA correlate well with those measured by FIA for serum with no significant differences of the mean concentration value for FSH (r=0.93, p: NS) and LH (r=0.99, p: NS), but with a significant difference for PRL (r=0.88, p=0.002). In CSF, mean concentrations values correlate well and we found no differences for TSH (r =0.97, p: NS) and LH (r = 0.94, p: NS), but significant differences for PRL (r= 0.98, p=0.001) and FSH (r=0.98, p=0.001). Statistically significant differences were also found for the CSF/serum ratio for PRL (p=0.08), FSH (p=0.001) and LH (p=0.001). The ratios CSF/serum are significantly higher with IRMA method than with FIA for all the hormones. Except for one patient for all the others we found the ratio CSF/serum less than 1 showing that the pathologic significance of this parameter is not modified due to the type of immunometric assay. A statistically significant difference (p<0.001) was found for mean FSH concentration in CSF with FIA and Chemiluminescent assays .\r\nConclusions: Any immunometric method currently used for the determination of hormones in the serum or plasma has to be validated accordingly in order to be used on CSF. For obtaining results that can be interpreted and compared it is preferred that the same immunometric method is used on both serum and CSF, inside one group of patients. The presence of a control group for the results determined with that method is strongly recommended.

As the kits used in routine for assessment of human growth hormone (hGH) status in CSF are originally designed for the serum and plasma, the specifications and limitations imposed by some environmental factor are important to be known. The aim of study was to\r\ninvestigate the influence of different protein concentrations on an immunometric assay to design a suitable and reproducible method for hGH measurement in different matrices, particularly in CSF-like environment.\r\nMaterial and methods. Experiments were performed using an in-house protocol for the determination of hGH by Time Resolved Immunofluorometric method. Standard curves were prepared in different biological matrices (sheep, human or fetal calf serums and CSFlike matrix). The concentrations of albumin and gamma-globulins were the variable parameters. The gold standard was the standard curve in sheep serum.\r\nResults. Constantly high background signal values were obtained at different albumin concentrations, in the absence of IgG, indicating intense non-specific binding and a low sensitivity. This fact allows accurate hGH measurement at over 5-10 ng/ml. Using a reaction environment enriched in IgG with a standard albumin concentration, background signal decreases while increasing IgG concentration. The highest sensibility is obtained for 5 g/l and 2.5 g/l IgG environments, allowing the sample signal to be measured accurately at hGH concentrations of 0.2 ng/ml or 0.5 ng/ml respectivly. This is observed in all biological matrices used.\r\nConclusions. The variation of protein concentration influences hGH determination. A standard curve with high sensibility for low values and a reduction in the influences of the reaction environment by minimising non-specific binding were obtained by enriching the environment with IgG.