LH and FSH in human cerebrospinal fluid.

LH and FSH of cerebrospinal fluid (CSF) and serum were radioimmunologically measured. Samples were obtained simultaneously from 116 subjects of the following groups: A. 22 patients with non-endocrine diseases, B. 18 patients with cranial diabetes insipidus (DI) of whom 4 with metastatic carcinomas, C. 5 patients with primary empty sella syndrome, D. one with hydatiform mole, and E. 70 patients with pituitary adenomas, i.e. growth hormone--or prolactin-secreting or "non-secreting" adenomas, of whom 38 patients with invasive and 32 with enclosed adenomas. LH and FSH are normal constituents of CSF and their CSF levels poorly correlates with the serum ones (LH r = 0.477 p less than 0.01). Enclosed adenomas with SSE showed low levels of LH in CSF. High CSF-gonadotropins concentrations (above 4.0 mIU/ml) with a low serum/CSF ratio (below 3) was frequently, but not constantly found in patients with invasive adenomas and are not indicative per se of this diagnosis. Some patients with brain metastasis from breast carcinoma and DI, or with non-tumoral diseases and DI showed similar high patterns of CSF gonadotropins though the serum levels were within normal range. This suggests that local vascular mechanisms, including the retrograde circulation of gonadotropins from the pituitary to the hypothalamus, influence the blood-brain barrier much more than the release of gonadotropins into the systemic blood circulation.

Schwangerschaftsprotein 1 (SP1) in serum and cerebrospinal fluid of patients with gestational trophoblastic disease.

The beta subunit of human chorionic gonadotrophin (beta-hCG) and Schwangerschaftsprotein 1 (SP1) were measured by radioimmunoassay in the serum and cerebrospinal fluid (CSF) of 46 postmolar and postpartum patients who developed gestational trophoblastic disease. There was a significant correlation between beta-hCG and SP1 serum levels. The mean serum SP1 level in high-risk patients was significantly higher than that in low-risk patients. There was a significant correlation between serum and CSF beta-hCG levels. The ratio of serum to CSF beta-hCG levels was low in the three patients with clinical evidence of intracranial metastasis. SP1 was present in the CSF of only one of these three patients, but it could be detected in the CSF of another four patients without clinical evidence of metastases in the central nervous system. The two low-risk patients with SP1 present in the CSF showed poor response to intramuscular methotrexate.

Cerebrospinal fluid human chorionic gonadotropin levels in normal pregnancy and choriocarcinoma.

We measured human chorionic gonadotropin levels simultaneously in the serum and cerebrospinal fluid of 36 normal pregnant women and of six patients with choriocarcinoma. During normal pregnancy, human chorionic gonadotropin levels in the cerebral spinal fluid correlated well with those in the serum in both the first and third trimesters. A serum-cerebral spinal fluid human chorionic gonadotropin ratio of less than 60 has previously been reported as a sensitive diagnostic test for trophoblastic central nervous system involvement. However, we found this ratio to be less than 60 in one patient undergoing first trimester abortion and in one patient with nonmetastatic choriocarcinoma. Furthermore, two patients with documented trophoblastic cerebral metastases had serum-cerebral spinal fluid human chorionic gonadotropin ratios in excess of 60. Results of our study, thus, indicate that determination of a single serum-cerebral spinal fluid human chorionic gonadotropin ratio may not conclusively prove or exclude trophoblastic central nervous system metastases. Unless it is corroborated by other diagnostic means, a single decreased serum-cerebral spinal fluid human chorionic gonadotropin ratio is insufficient evidence to initiate multimodality treatment for presumptive central nervous system involvement.