| ObjectiveIncidence of pituitary tumor in reproductive age female is getting higher and higher, but comprehensive analysis of a large sample is scarce. We analyzed characteristics of387reproductive age females with pituitary tumor,including clinical manifestation,image features.pathologic results,effect of operations,further treatment of residual or recurrent tumor and postoperative complications to further guide on the clinical diagnosis, treatment and follow-up of pituitary tumor and improve prognosis. In addition, gonadal dysfunction is a major problem that plagued reproductive age females with pituitary tumor. Through the analysis of the influence factors of gonadal dysfunction, we can manage earlier for women who desire to fertility or restore normal menstrual cycle and also delay the advent of complications,for example osteoporosis.cardiovascular disease and so on.MethodThe medical records of387women with pituitary tumor who were from15to50years old, had been treated in the Shanghai fifth people’s hospital from January1st2003to Marth thirty-first2012and had complete clinical data were collected.Firstly, we summary and analyze distribution of age and size,first clinical manifestations,image features, pathologic results. treatment means, the common tumor types characteristics and relationship between preoperative hormone levels and immunohistochemical results:What’s more, we assess incidence and predisposing factors of postoperative complications, such as electrolyte imbalance, polyuria or diabetes insipidus and fever; Finally, we were follow-up about clinical manifestations, hormone replacement,hormone level, percent of remission, residual and recurrence, gonadal function and risk factors of gonadal dysfunction. Statistical analyses were performed SPSS software17.0. The test including Independent sample T test,variable transformation T test and nonparametric rank sum test (Mann-Whitney U test) method is used to compare continuous variables for two groups. But univariate analysis of variance (ANOVA) is used for several groups.Relation analysis use Spearman and partial correlation analysis; Analysis of categorical variables use Pearson X2, Fisher exact test. Paired Chi-square test and logistic regression Result387patients were diagnosed pituitary adenoma before surgery.69.5%patients were macroadenomas and60.21%patients were above30years old.Tumor size was positively related with patient’s age(r=0.355, P=0.000).92.5%patients complained of the symptoms caused by excessive hormone,including menstrual disorder,galactorrhea, infertility, acromegaly and so on. The younger the patients, the higher preoperative GH or PRL levels, the more the patients were diagnosed by endocrine manifestation. The common endocrine manifestations were menstrual disorder, menopause and galactorrhea and the usual mass effect were headache and vision loss.The most frequent pathological type was prolactin(PRL)-secreting adenoma(34.1%) in child-bearing age, followed by growth hormone(GH)-secreting adenoma(39,10.1%), PRLand GH-secreting mixed tumor (31,8.0%), nonfunctioning adenoma(NFPA)(29,7.5%), adrenocorticotropic hormone(ACTH)-secreting adenoma (23,5.9%),plurihormonal pituitary adenomas(19,4.9%). Macroadenomas is the most common type. In addition to PRL adenoma, the majority of patients of others types of pituitary adenomas is greater than30years old.58.7%patients were invasive pituitary adenomas, the rest of patients(41.3%) were non-invasive pituitary adenomas.Invasive tumor was related with preoperative PRL levels (OR=1.004,P=0.000), tumor size (OR=1.121,P=0.000) and clinical manifestation by preoperative excessive hormone (OR=24.25,P=0.037). The patients with invasive tumor had higher PRL level,larger tumor size and no clinical manifestation of excessive hormone.Median preoperative PRL levels and tumor size ware88.8ug/L and20mm,respectively.Comparing with non-PRL tumor, the patients with PRL tumor were smaller tumor size.younger and higher preoperative PRL levels. Median were12mm,28years old181.9ug/L, respectively.Preoperative PRL level was positively correlated with tumor size (r=0.242, P=0.013),but negatively correlated with age(r=-0.288,P=0.003).Preoperative PRL level diagnosis cut-off point value was112.2ug/L(sensitivity78%, specificity83%). Prognostic factors in prolactin pituitary tumors of reproductive age female were tumor size, preoperative PRL level, PRL level of the first postoperative day, reduction rate of PRL level of the first postoperative day and invasive tumor. Remission rate of the patients which postoperative PRL level was less than23.46ug/L. reduction rate of PRL level was above90.78%, preoperative PRL level was less than260ug/L,tumor size was less than14.5mm were66,27.75,8.54and14.92times, respectively. Remission rate of non-invasive tumors is3.76times of invasive tumors. Remission rate was higher after transsphenoidal surgical treatment than that reported in the literature medication.Comparing with non-GH tumor. the patients with GH tumor were higher preoperative GH level, larger tumor size and older.Median were14.3ug/L,22.5mm and41years old,respectively.There were no correlation between preoperative GH level and tumor size. Preoperative GH level diagnosis cut-off point value was2.6ug/L (sensitivity77%, specificity83%).All of ACTH adenomas, Silent corticotroph adenomas(SEA) accounted for91.3%. Active ACTH tumors accounted for only8.7%.Pateints with ACTH tumors were larger tumor size and older than non-ACTH tumors. Median tumor size and age were26.4mm and37.6years, respectively.The levels of ACTH and F was no difference between ACTH tumors and non-ACTH tumors.6patients were diagnosed as active ACTH tumors before surgery.However, pathological ACTH-positive were only2cases. Preoperative ACTH and F level for diagnosing ACTH tumor are no value.72.4%of the non-functioning adenomas and94.4%of follicle stimulating hormone(FSH)-secreting tumors associated with hyperprolactinemia. Average PRL levels in FSH tumors was74ug/L.Comparing with non-FSH tumors, Patients with FSH tumor were higher FSH levels.When imaging tests proved pituitary tumor and PRL level slightly elevated. Preoperative FSH level diagnosis cut-off point value was6.685mIU/mL (sensitivity72.2,%, specificity79%).369patients (95.3%) with pituitary tumor were treated with surgical treatment by the same expert surgeon, of which352cases (95.4%) through the sphenoid sinus resection,7cases (1.9%) through the keyhole approach,7cases (1.9%) craniotomy, the rest of surgical approach is unknown.Preoperative secondary adrenocortical insufficiency and normal were61(16.8%),87cases (23.9%),respectively. Adrenocortical function were not judged and should give a stimulating test in216cases (59.3%). Secondary hypothyroidism, primary hypothyroidism (Hashimoto’s thyroiditis), normal thyroid function and unknown thyroid function were22(6%),6(1.6%),253(69.3%) and83cases (22.7%) respectively. Indeed,women of reproductive age, assessing the menstrual status is the best means to make the diagnosis of gonadal dysfunction. So normal gonadal function and hypogonadism were57patients (14.73%),321cases (82.95%),respectively before surgery. In menstrual disorder’s patients, we based on hormone levels to determine preoperative hypogonadism only145cases (37.5%).243(62.79%)patients were follow-up, of which227patients were treated with surgery,40patients (18%) were not follow-up any of anterior pituitary hormones.67patients (73%) were follow-up at least one anterior pituitary hormones and the rest of20patients (9%) were unknown.The proportion of LH and FSH insufficiency were7.5%,ACTH,TSH,GH and PRL insufficiency accounted for5.3%.4.4%,3.1%and1.8%,respectively.The common of postoperative complications in women of reproductive age were hyponatremia (64,17.3%), hypernatremia (77,20.9%), fever (19,5%), and polyuria (264,71.5%).Craniotomy which caused postoperative serum sodium<130mmol/1or greater than150mmol/l level is19.7times of transsphenoidal surgery.24hours urine volume of66.9%patients were distributed between3000and5000ml after surgery.The greater the age, the higher incidence of postoperative polyuria was.(OR=1.032, P=0.032).The common of clinical symptoms were hyposmia (22.47%), and fatigue (19.82%), sexual dysfunction (16.74%), pubic hair, armpit hair or beard off (14.54%), cold (12.78%) and constipation (7.49%) during follow-up.There were137(61%),12(5%),41(18%) and2(1%) patients in remission, relapse, residue and death, respectively. The patients who were the degree of tumor invasion of Hardy V-class was0.4times of non-V-class by multivariate analysis.321patients (82.9%) were menstrual disorder,57(14.7%) patients were menstrual normal and9patients (2.3%) had gynecological disease. Comparing with preoperative non-menstrual disorder group, menstrual disorders group had higher PRL level, lower E2levels and younger by univariate analysis. Median were98.36ug/L,7.43pg/mL,32years old, respectively.However,multivariate analysis just showed that preoperative PRL level and age were related with menstrual status. Corelation analysis showed that preoperative PRL lavel was negatively correlated with E2level(r=-0.172,P=0.001).63.39%(142) patients were normal menstruation and32.59%(73) patients were menstrual disorders during follow-up. Paired chi-square test showed that patients can recover normal menstrual cycle after surgery (P=4×10-3). The patients who were ACTH tumor, PRL tumor, no previous pituitary tumor treatment history, smaller tumor size had easier normal menstrual cycle after surgery by multivariate analysis. OR values were1.205,1.19,2.7.1.09,respectively.In the patients with preoperative menstrual disorder, the patients of macroadenomas and prolactin normal or reduction (Hu and NPRL) were harder to recover menstrual cycle than which of macroadenomas and above normal PRL level (Hu and HPRL)(OR=0.13,95%CI:0.03-0.48).The patients of Hu and HPRL were also harder to menstrual cycle than which of microadenoma and above normal PRL level (Sm and HPRL)(OR=0.34,95%CI:0.13-0.89).ConclusionThe most incidence is at age above30years old or macroedenoma. Tumor size is positively correlated with age.Most of pituitary tumors dued to manifestation of excessive hormone. The younger the patients, the higher preoperative GH or PRL levels, the more the patients were diagnosed by endocrine manifestation. The most common tumor type is Invasive pituitary tumor. The patients with invasive tumor had higher PRL level, larger tumor size and no clinical manifestation of excessive hormone. The most frequent pathological type was PRL tumor, followed by GH tumor.The majority of PRL tumors patients were macrodenoma or less than30years old.Preoperative PRL level was positively correlated with tumor size (r=0.242, P=0.013),but negatively correlated with age(r=-0.288,.P=0.003). Preoperative PRL level diagnosis cut-off point value was112.2ug/L. Prognostic factors in prolactin pituitary tumors of reproductive age female were tumor size, preoperative PRL level, PRL level of the first postoperative day, reduction rate of PRL level of the first postoperative day and invasive tumor,of which PRL level of the first postoperative day less than23.46ug/L is the most prognosis factor. Remission rate was higher after transsphenoidal surgical treatment than that reported in the literature medication.The majority of GH tumors patients were macrodenoma or above30years old.Comparing with non-GH tumor. the patients with GH tumor were higher preoperative GH level, larger tumor size and older.Preoperative GH level diagnosis cut-off point value was2.6ug/L. The most common of tumor type is SEA in ACTH adenomas. The majority of SEA were macroadenomas.Activity ACTH tumors are scarce and microadenoma.Sometimes pathological ACTH-negative can be presented in activity ACTH tumor. Preoperative ACTH and F level for diagnosing ACTH tumor are no value.The mojority of non-functioning adenomas and FSH tumor were macroadenoma and associated with hyperprolactinemia.Incidence of postoperative polyuria, hyponatremia and hypernatremia were high. The greater the age, the higher incidence of postoperative polyuria.Incidence of serum sodium irregularity was higher though Craniotomy. Only one time hormone test was not to assess secondary hypoadrenocorticism and hypogonadism.The stimuation test should be executed.The degree of tumor invasion which is not up to the Hardy V-class is a good index to predict postoperative remission rate.Predictive factors of preoperative normal gonadal function were preoperative PRL level and age. The mechanism which preoperative PRL induce gonadal dysfunction was reduce E2level. Patients can recover gonadal function after surgery.The patients who were ACTH tumors, PRL tumors, no previous pituitary tumor treatment history, smaller tumor size had easier normal gonadal function after surgery. In the patients with preoperative gonadal disorder, the patients of Hu and NPRLwere harder to recover gonadal function than which of Hu and HPRL.The patients of Hu and HPRL were also harder to gonadal function than which of Sm and HPRL. |