ACTA ENDOCRINOLOGICA (BUC)

accepted. Such an intrinsec system of the pineal gland was revealed by Rumanian scientists, which found a high renin like biological activity in the rat pineal, after nephrectomy, and also de novo production of isorenin by the monolayer cultured pineal cells; angiotensin I (A I) - immunoreactivity -was also found in the rat pineal at much higher concentration than in the brain structures. In this paper we add morphological evidences regarding components of the RAS in the bovine pineal gland. Bovine pineals were processed by immunohistochemistry for angiotensinogen, angiotensins I and II, as well as for renin. Frozen pineals were processed by "in situ" hybridization for angiotensinogen and renin mRNAs. Angiotensin converting enzyme(ACE), aminopeptidase A and angiotensin III (AIII) were not investigated. Our data showed that a number of pinealocytes stains for angiotensins (AI and AII), but the renin staining was negative. No nerve fibres were stained with AI or AII. The angiotensins-positive cells were dispersed in the pineal parenchyma, without a particular distribution in relation to the blood vessels. The angiotensinogen immunohistochemistry showed sparse positive cells. The "in situ" hybridization data showed the signal for angiotensinogen mRNA located over pinealocytes, while the renin mRNA was absent. The local synthesis of angiotensinogen in the bovine pineal cells, as a substrate for generation of AI and AII are supported by our data. However, at least the first enzyme involved in this cascade is different from the typical renin. The presence of an angiotensin system without renin in the pineal gland is an additional argument to the evidence that angiotensin II and related peptides can be formed by renin- and ACE- independent protease pathways.

We highlight here in a human pituitary prolactinoma, by double immunohistochemistry, clusters of cells with strong reactivity for Sox2 and prolactin The colocalization was mainly distributed at the periphery of the tumor. In contrast, in the central part of the same tumor, only\r\nisolated cells expressed Sox 2 colocalised with prolactin. To the best of our knowledge, this is the first microscopic evidence of Sox2 positivity in prolactin secreting cells from human pituitary and from such type of human pituitary adenoma.

Right ventricular dysfunction (RVD) is critical for risk stratification of patients with pulmonary embolism (PE). Evaluation can be made by echocardiography or biological markers among which plasma levels of brain natriuretic peptide (BNP). The aim of our study is assessment of BNP levels in patients with PE associating or not RVD as diagnosed by classic echocardiographic criteria. We prospectively assessed 40 patients with deep venous thrombosis and confirmed PE (age range 52.5 ? 9.14 years, 22 men and 18 women), with (14) or without (26) RVD on echocardiography. Plasma BNP levels were significantly higher in RVD patients (190 ? 171.2 pg/mL vs. 15.75 ? 18.85 pg/mL, P < 0.0001). A cut-off level of plasma BNP = 50 pg/mL had a sensitivity, specificity, positive and negative predictive value for the diagnosis of RVD of 84% (C.I. 79% - 88%), 80% (C.I. 75% - 85%), 83% (C.I. 77% - 87%) and 79% (C.I. 75% - 84%), respectively. There was a significant correlation between plasma BNP levels and end-diastolic RV diameter (r = 0.56, P < 0.0001), RV systolic pressure (r = 0.50, P < 0.001) and the presence of a Qr complex in V1-lead on ECG (r = 0.55, P < 0.05). Four patients with RVD on echocardiography and syncope, all admitted relatively soon after the onset of their symptoms, had BNP in normal range. In conclusion, PE should be considered in the differential diagnosis of patients with dyspnoea and increased plasma BNP levels. A cut-off level of 50 pg/mL could identify the RVD in patients with PE with a good sensitivity and specificity. Normal range plasma BNP levels do not exclude even a severe PE and should be interpreted with caution, especially in highly symptomatic patients with recent symptom onset.

controversial. Aim. This study retrospectively reviews the tumor evolution in patients with NFA macroadenomas treated with surgery and conventional RT, as compared to surgery alone. Methods. Of 107 unselected patients with operated NFA (aged 19 - 77 years), evaluated between 1977 - 2008, 71 patients were follow-up without RT (group A), while 36 patients were submitted to RT (group B). Patients submitted to radiosurgery were not included. Both groups underwent serial imaging studies with computed tomography or magnetic resonance. Tumor evolution was conventionally defined as a change of minimum 25% of diameter. Results. The surgical approach was transfrontal in 47% of patients, transsphenoidal in 43% or both in 10% of patients, similar in both study groups. In group B, 30 patients underwent highvoltage RT (mean total dose 50.5 Gy) and 6 patients low-voltage RT (mean emission dose 16,775 R). Mean follow-up after surgery in group A was 3.4 years (range 6 months - 10 years) and after RT in group B it was 6.8 years (range 2 &#8211; 24 years), p < 0.001. In group A, 16 out of 71 patients had no visible tumor remnants. In this subgroup, 2 patients (12.5%) showed tumor recurrence. Fifty-five out of 71 patients had residual tumors, 21 with extrasellar extension after surgery. In this subgroup, 21 patients (38%) showed tumor re-growth and 7 (13%) showed tumor decrease. In group B (n=36) all patients had tumor remnants after surgery with extrasellar extension in 30 patients. After RT, tumor re-growth occurred in 5 out of 36 patients (14%) as compared to subgroup A with residual tumors (p< 0.05) and tumor decrease in 14 out of 34 (41%), as compared to the same subgroup A (p < 0.01). The 5 year-tumor re-growth free survival rate of 88% in irradiated patients was significantly better than in non-irradiated patients with residual tumors (31%, log-rank test, p < 0.01, Kaplan-Meier analysis), but similar to that in patients without remnants (87.5%). Age, sex, tumor parasellar extension and size of residual tumor were not predictors of recurrency. Conclusion. Postoperative radiotherapy provides a significant improvement of local control in patients with residual NFA compared to surgery alone. It is necessary a long term follow-up due to recurrency noticed up to 8 years postsurgery. In patients without tumor remnants, a wait-and-see policy is indicated after surgery.

Background. Hirsutism is part of current criteria of polycystic ovary syndrome (PCOS), as a clinical expression\r\nof hyperandrogenism.\r\nObjective. To evaluate the significant factors for hirsutism severity in PCOS.\r\nPatients. A total of 235 PCOS patients, consecutively coming for medical advice, aged 18-35 yrs, all of Romanian origin, were diagnosed according to Rotterdam criteria.\r\nMethods. Hirsutism, quantified using the modified Ferriman-Gallwey (mFG) procedure, was defined by values equal or\r\nmore than 6. Other parameters evaluated were: body mass index (BMI), fasting insulinemia, insulin resistance quantified by QUICKI, total testosterone (TT), free androgen index (FAI), dehydroepiandrosterone sulfate (DHEAs), 17OH progesterone, fasting glycemia. In a subset of 106 patients, androgen receptor (AR) was explored by CAG repeat\r\ngenotyping and X-chromosome inactivation analysis.\r\nResults. The total PCOS population (235) was divided in group A (n=139, 59.14%) with hirsutism and group B (n=96, 40.85%) without hirsutism. In univariate correlations, serum\r\ninsulin levels (p<0.05) and insulin resistance quantified by QUICKI (p<0.05), but not FAI, TT, DHEAs, 17OH progesterone or BMI were associated significantly with mFG score, in group A of hirsute PCOS patients and also in group B\r\nof nonhirsute PCOS. In a stepwise regression mFG model,\r\nincluding TT, insulin and BMI, only insulin remained independently associated with mFG score (p<0.05) in the group A of hirsute PCOS patients, whereas in group B\r\nof the nonhirsute PCOS, there were not significant associations. Androgen receptor parameters explored in 106 cases, i.e. by the biallelic means and X-weighted biallelic means of CAGn, did not show significant associations with mFG score in univariate correlations. Only insulin was significantly associated (p<0.05) in another stepwise\r\nregression model of mFG including as parameters insulin, TT, biallelic means of CAGn and BMI.\r\nConclusions. Our results support that insulin is significantly associated with the\r\nseverity of mFG score in PCOS patients, independent of serum androgens or androgen receptor sensitivity expressed by\r\nCAGn polymorphism. This suggests a possible pathogenic role of high insulin level for the development and progression\r\nof hirsutism, at least in PCOS.

In this study we aimed to assess amiodarone&#8217;s effects on thyroid hormones, C reactive protein (CRP), interleukin 6 (IL-6) and tumor necrosis factor (TNF-&#945;) profile in amiodarone-induced thyrotoxicosis (AIT) and their relationship with arrhythmias. In 60 patients with AIT (29M/31F), aged 59.7 ? 1.5 years and 105 hyperthyroid patients (25M/80F), aged 59.4 ? 1 years, TSH, total T3, total T4, free T4 were measured by IRMA or microenzymatic immunoassay. In 11 AIT patients and 26 patients with common hyperthyroidism, without significant Graves&#8217; ophthalmopathy, CRP, IL-6 and TNF-&#945; were measured by chemiluminescent immunoassays. AIT patients showed significantly lower T3 levels (273.82 ? 18.76 ng/dL) as compared to common thyrotoxicosis (361.89 ? 13.47 ng/dL), p<0.001, while T4 and FT4 were similar. AIT patients showed similar CRP, IL-6 and TNF-&#945; levels to common hyperthyroidism (10.22 ? 3.03 mg/L vs. 6.31 ? 2.33 mg/L, p=0.11, for CRP, 17.46 ? 13.09 pg/mL vs. 6.46 ? 1.14 pg/mL, p=0.65, for IL-6, and 9.9 ? 1.76 pg/mL vs. 12.85 ? 2.22 pg/mL, p=0.22, for TNF-&#945;, respectively. Patients with arrhythmias (n=19) showed significantly higher CRP, IL-6 and TNF-&#945; levels as compared to patients without arrhythmias (n= 18): 10.81 ? 3.35 mg/L vs. 3.96 ? 1.08 mg/L, p=0.05, for CRP, 15.15 ? 7.52 pg/mL vs. 4.02 ? 0.74 pg/mL, p=0.007 for IL-6, and 15.16 ? 3.33 pg/mL vs. 9.2 ? 0.82 pg/mL, p=0.037 for TNF-&#945;. In conclusion, AIT showed a similar pattern of proinflammatory cytokines to common hyperthyroidism. Increased CRP, IL-6 and TNF-&#945; are found only in patients with thyrotoxicosis associated with atrial fibrillation or other arrhythmias.

This study evaluates the differences of radiotherapy in patients with pituitary tumors, in relation to gonadotropin immunoreactivit.\r\nDesign. It is a longitudinal, retrospective study of 117 patients submitted to pituitary surgery and high voltage radiotherapy. The excised tumors were 70 non-functioning adenomas (NFA) and 47 GH-secreting adenomas producing active acromegaly (ACM). They were evaluated before and after pituitary surgery, before radiotherapy as baseline, then at 3 different intervals at 0 - 2, 2 - 5 and > 5 years after baseline.\r\nMethods and patients. Computer tomography was used for measuring the tumor size and specific immunoassays were used for FSH, LH and nadir GH during 75 g oral glucose load. Immunohistochemistry (IHC) was performed with avidin-biotine method. High voltage conformational radiotherapy used a linear accelerator of 10 meV, with a 50 Gy on target tumor. For statistics, student&#8217; t test was used. Data before surgery (tumor volume and hormonal sexretion) were available in 70 unselected patients (31 NFA and 39 ACM from the above group). Postsurgery we defined following groups: NFA-A1 exposed to radiotherapy (n=21) and NFA-C1 unexposed to radiotherapy (n=22); ACM-A2 exposed to radiotherapy (n=20) and ACM-C2 unexposed to radiotherapy (n=10).\r\nResults. Immunohistochemistry for NFA showed 27 immunopositive for FSH or/and LH (GD+) and 40 immunonegative for FSH and LH (GD-), 3 undetermined, while for ACM were 12 GD+, 33 GD-, 2 undetermined. Immunohistochemistry data on defined groups was as follows: NFA-A1 (n=21: 12 GD+, 9 GD-) and NFA-C1 (n=22: 6 GD+, 16 GD-); ACM-A2 (n=20: 4 GD+, 16 GD-) and ACM-C2 (n=10: 3 GD+, 7 GD-). In patients with NFA presented before therapy, there are not significant differences of tumor sizes or of the levels of FSH/LH between GD+ and GD- adenomas. In ACM, before any therapy, the GD+ patients showed a significantly higher FSH levels (20.7+11.4 U/L, n=6) than GD- patients (FSH 6.6+1.6 U/L, n=22, p< 0.05) and a nonsignificant lower serum GH levels (15.1+3.5 ng/mL, n=8 versus 33.5+8.9 ng/mL, n=30 p=0.06), although the tumor size was similar between the two groups. Radiotherapy upon NFA: GD+ adenomas did not decrease their volume after radiotherapy (cranio-caudal diameter 1.63+0.79 cm, before and 1.54+0.68 cm at 2 - 5 years post-radiotherapy n=6, p= NS), in contrast with GD- tumors in which a slightly, but significant decrement in volume could be demonstrated (from 2.79+0.53 cm to 2.43+0.31 cm at 2 - 5 years, n=5, p= 0.01). Radiotherapy in ACM resulted in a decrement of serum GH level and tumor size, as compared with the control group without radiotherapy. The effect was maximal at the interval of 2-5 years. The ACM, GD- tend to respond better to radiotherapy, (i.e. GH levels decreased from 15.1+5.4 to 6.6+2.4 ng/ml at 2-5 years, p=0.05), while in patients with ACM, GD+ the GH level did not show a significant decrease (serum GH was 7.3+3.3 ng/ml before and 5.1+4 ng/mL at 2-5 years post-radiotherapy, p = NS). The CC diameter of GD- decreased from 1.1+0.3 to 0.7+0.2 at 2-5 years, p=.059, while in GD+: from 1.64+0.4 to 1.2+0.3 ng/mL at 2-5 years, p = NS.\r\nConclusion. Pituitary adenomas, both NFA and ACM that contain gonadotropin immunoreactive cells tend to be more radioresistant than those without gonadotroph cells.

Aim: We aimed to assess the final outcome of combined therapeutic approaches in patients with macroprolactinomas that were resistant to dopamine agonists (DA).\r\nPatients: Records of patients with macroprolactinoma hospitalized in the Institute of Endocrinology, Bucharest, between 1978-2005, were reviewed. There were 29 eligible patients resistant to DA therapy (8 men and 21 women), out of 119 patients with macroprolactinomas (24.4%); age at diagnosis of the resistant patients ranged between 16-59 years (31.9 ? 2.4 years), with mean prolactin (PRL) levels 2,110.2 ? 656.6 ng/mL (range 42-16,000 ng/mL). The mean maximal tumor diameter was 2.7 ? 0.2 cm (range 1-6.8 cm).\r\nMethods: Rapid fluoroimmunoassay using Europium was used for hormonal levels; computed tomography imaging and/or MRI were used to assess tumor size. Study design: The resistance to DA drugs was evaluated using initial criteria: the lack of prolactinoma response to current daily dose of Bromocriptine (BRC) 7.5 mg/day or to Cabergoline (CAB) up to 2 mg/week for at least 6 months (step 1) or final criteria: the lack of response to high BRC doses (30 mg/day) or CAB doses between 2.5-4 mg/week for at least 6 months (step 2). The lack of response was considered if PRL levels remained above the upper normal limit (20 ng/mL) and the tumor mass size decreased by less than 50%. All resistant cases at step 1 received thereafter maximal medical therapy with DA drugs, according to step 2. Thereafter, resistant macroprolactinomas after step 2 were submitted to step 3 - high voltage radiotherapy ? surgery. Serum PRL levels and tumor size were finally evaluated 110 ? 26.5 months later (range: 6-381).\r\nResults: Outcome of medical therapy with DA (n=29): Overall, 7 out of 29 resistant macroprolactinomas (24.1%) were successfully treated by increasing BRC dose (n=5) or changing BRC to CAB (n=2). But 22/119 (18.5%) patients remained resistant to DA drugs independent of dose, duration or type of drug used. 14 patients failed to normalize PRL levels despite CAB treatment in doses up to 7 mg/week. Outcome of radiotherapy alone or combined with surgery (n=15): PRL normalization was achieved in 4 patients out of the only 7 assessed at least at 18 months after radiotherapy. Withdrawal of DA therapy revealed 2 cured cases, both after radiotherapy and surgery. Outcome of surgery: Only one patient normalized PRL levels after surgery, but she soon relapsed. Apparently, only one case of acquired resistance to DA drugs was revealed. We found that 15.1% (18/119) of the patients with macroprolactinoma did not finally normalize their serum PRL even after combined therapy approaches (DA + radiotherapy ? surgery), after 79 ? 17.4 months (range 6 to 206 months) treatment total duration and 45.4 ? 19 months (range 3 to 206 months) after radical therapies, respectively.\r\nConclusion: In summary, the resistance was successfully treated in 38% cases (11 out of 29).

Today therapeutic options for obesity include miniinvasive methods. The authors present the results of a prospective study regarding a lot of 224 obese patients treated between January 2003 ? January 2005 with laparoscopic adjustable gastric banding (LAGB) (n=184) and endogastric balloon (n=40). The medium body mass index (BMI) of admission for LAGB was 46.08 kg/m2 (extremes between 36 ? 71 kg/m2), 42 patients being super obese (BMI over 50 kg/m2). There were no conversions to open techniques. Postoperative evaluations were at 1, 6 and 12 months. Medium BMI was 39.1 kg/m2 (115 patients) at six months after surgery and 38.78 kg/m2 (59 patients) at one year after surgery. Comorbidities were remitted at 72% of patients at 6 months postsurgery. The patients treated with endogastric balloon endoscopically introduced (40 cases) were followed 6 months, after that they were extracted (26 cases). The medium BMI of admission was 33 kg/m2 (extremes between 21 ? 43 kg/m2) and at 6 months the medium BMI was 26.8 kg/m2. Although we do not benefit of a long time follow-up, the favorable initial results allow us to state that miniinvasive techniques deserve an important place in the efforts of struggling against obesity and its consequences.