Pituitary hemorrhage in pituitary adenomas treated with gamma knife radiosurgery: incidence, risk factors and prognosis

Objective: The aims of this study were to investigate the incidence, risk factors and prognosis of pituitary hemorrhage in pituitary adenomas treated with gamma knife radiosurgery (GKRS). Methods and materials: Between December 1993 and December 2016, 751 consecutive pituitary adenoma patients treated with GKRS were retrospectively reviewed in a single center. There were 271 male (36.1%) and 480 female (63.9%) patients with a median age of 38.5 (range, 7.2-84.0) years. The number of nonfunctioning pituitary adenomas (NFPAs) and functioning pituitary adenomas were 369 (49.1%) and 382 (50.9%) respectively. The median follow-up time was 61.1 (range, 12.1-304.4) months. Results: In this study, 88 patients (11.7%) were diagnosed with pituitary hemorrhage before GKRS, 55 patients (7.3%) developed new or worsened pituitary hemorrhage after GKRS (excluding 3 patients with new or worsened pituitary hemorrhage due to tumor regrowth). The median time to new or worsened pituitary hemorrhage after GKRS was 18.9 (range 3.1-70.7) months. Overall, 128 patients (17.0%) were diagnosed with pituitary hemorrhage in the entire series. After adjustment with logistic regression, nonfunctioning pituitary adenomas (NFPAs) (odds ratio [OR]=2.121, 95% confidence interval [CI]=1.195-3.763, p=0.010) and suprasellar extension (OR=2.470, 95% CI=1.361-4.482, p=0.003) were associated with pituitary hemorrhage before GKRS. NFPA (OR=3.271, 95% CI=1.278-8.373, p=0.013) was associated with new or worsened pituitary hemorrhage after GKRS. Five patients received surgical resection for new or worsened pituitary hemorrhage were considered as GKRS treatment failure. Two patients with new hypopituitarism were considered to be owed to new or worsened pituitary hemorrhage after GKRS. Conclusions: New or worsened pituitary hemorrhage after GKRS was not an uncommon phenomenon. NFPA was an independent risk factor of new or worsened pituitary hemorrhage after GKRS. New or worsened pituitary hemorrhage after GKRS could lead to GKRS treatment failure. GKRS might be a precipitating factor of pituitary hemorrhage.


Introduction
Pituitary adenomas are one of the most frequent intracranial primary tumors, representing about 16.7% of all brain tumors [1]. These adenomas can be classified as nonfunctioning pituitary adenomas (NFPAs) and functioning pituitary adenomas (FPAs) according to the evidence of secreting biologically Ivyspring International Publisher active hormones.
Clinical pituitary apoplexy is a rare disorder with an acute clinical syndrome characterized by severe headaches, sudden visual impairment and ophthalmoplegia due to hemorrhage or infarction of pituitary adenomas and related with significant morbidity. It may present more common in NFPAs than FPAs [2]. Some epidemiological studies reported its incidence was 0.17 episodes per 100 000 per year [3] and 6.2 cases per 100 000 inhabitants [4]. Almost 0.6%-10.0% of pituitary adenoma patients experience apoplexy [2,[5][6][7]. Comparing to acute apoplexy, subclinical apoplexy based on radiological findings or autopsy which is asymptomatic, is more common, with up to 25% of pituitary adenomas [8][9][10][11][12][13]. Pituitary hemorrhage is the main form of clinical or subclinical pituitary apoplexy, with up to 28.1% in pituitary adenomas [14].
Predisposing and precipitating factors of pituitary hemorrhage including arterial hypertension, intracranial pressure, major surgery, closed head trauma, dynamic tests, angiography, dopamine agonists (DAs) and anticoagulant therapy had been widely reported [9]. Radiation-induced pituitary hemorrhage had been reported previously. But only a small number of cases were reported thus far [15][16][17][18][19]. In our previous study [20], the early pattern of treatment failure in initial Gamma Knife radiosurgery (GKRS) for NFPAs was pituitary apoplexy. Seven patients (8.6%) developed new pituitary hemorrhage after GKRS. Tumor volume (≥10 cm 3 ) was an independent risk factor. GKRS has been the most commonly used radiation technique and essential part in the treatment of pituitary adenomas, with advantages of better dose conformity, a highly precise and focused delivery of radiation in a single session. However, little is known about the incidence, risk factors and prognosis of pituitary hemorrhage in pituitary adenoma patients treated with GKRS so far.
There are more than 26 years' experience using GKRS (Elekta, Stockholm, Sweden) for pituitary adenomas at the Second Affiliated Hospital of Guangzhou Medical University since 1993. In order to report the incidence, risk factors and prognosis of pituitary hemorrhage in pituitary adenomas treated with GKRS, we performed a large single-center retrospective study.

Patient population
Between December 1993 and December 2016, the medical records of 2557 patients who underwent GKRS for the treatment for pituitary adenomas were retrospectively reviewed at the Second Affiliated Hospital of Guangzhou Medical University. 1806 cases of patients who were absence of clinical data and follow-up information in our hospital were excluded. Finally, 751 patients with available clinical data and at least one follow-up were included in this study. This retrospective study was approved by the institutional committee of the Second Affiliated Hospital of Guangzhou Medical University.

Definition of pituitary hemorrhage in pituitary adenomas
In this study, pituitary hemorrhage was diagnosed according to magnetic resonance imaging (MRI), computed tomography (CT), surgical findings or pathological findings. All of patients had MRI of sellar before and after GKRS. Images were reviewed by an experienced radiologist to confirm pituitary hemorrhage. The diagnosis of pituitary hemorrhage in MRI findings was mainly based on the following signs [21][22][23]: (1)

Clinical and radiological evaluations
Patients were required to take routine clinical and radiological follow-up evaluations every 6 months for the first 3 years and thereafter yearly. Clinical pituitary apoplexy was diagnosed based on episode of acute severe headaches, sudden visual disturbance or opthalmoplegia as well as presence of a pituitary adenoma. Subclinical pituitary apoplexy was defined as an intratumoral hemorrhage based on MRI, CT, surgical or pathological findings, with no clinical symptoms of acute pituitary adenoma apoplexy. Tumor dimensions were obtained manually from MRI. The dimensional indices of the tumors were measured and recorded in three orthogonal planes: transverse (TR), anteroposterior (AP), and craniocaudal (CC). Tumor volumes were estimated using the following formula: V = (π × [TR × AP × CC])/6 [24]. Parasellar invasion was defined as Knosp grade 3 or 4. Suprasellar extension was defined as the presence of a tumor close to optic nerve and chiasm (< 2mm). Predisposing factors of pituitary hemorrhage, including hypertension, diabetes mellitus, DAs, anticoagulants, and angiography, were reviewed in the medical records for all patients.
Hypopituitarism was defined asymptomatic anterior pituitary deficits with laboratory documentation. New hypopituitarism was defined by requiring hormone replacement or a new deficiency in any one of the hormonal axes after GKRS. Hypothyroidism was based on low free thyroxine (normal range, 12.00-22.00pmol/L). Hypocortisolism was defined if morning cortisol level (08:00) was <100 nmol/L. If morning cortisol level was >400 nmol/L, hypocortisolism was excluded. Hypogonadism in males was defined by low serum testosterone levels (<12 nmol/L) without elevated luteinizing hormone (LH) and follicle stimulating hormone (FSH) levels. Hypogonadism in females was defined by amenorrhea with low serum estradiol and low gonadotrophins in premenopausal women, and the absence of high gonadotrophins (LH and FSH) in menopausal women.

Gamma knife radiosurgery technique
All of patients underwent GKRS in our hospital. The procedure was performed using model B Leksell Gamma Knife Unit before April 2014 and Perfexion Unit (Elekta Instrument, Inc. Stockholm, Sweden) thereafter. Following stereotactic Leksell frame placement, thin-slice stereotactic MRI with contrast was performed through the sellar. Subsequently, GKRS treatment planning was made in consultation with a medical physicist, radiation oncologist and neurosurgeon. Dose selection for GKRS was mainly dependent on the tumor type, tumor volume, distance to the optic nerve and chiasm. The median tumor margin dose was 14.4 Gy (range, 8.0-30.0 Gy) at a median prescription isodose 40% (range, 18.8-80.0%). The median maximum dose was 33.3 Gy (range, 14-90 Gy).

Statistical analysis
The normal distribution of continuous variables was checked by Kolmogorov-Smirnov test. The median and interquartile ranges were used to describe variables not normally distributed. When continuous variables were not normally distributed, Wilcoxon rank sum test was used. Chi-square test and Fisher exact test were used for statistical analysis of categorical variables. Logistic regression was performed using variables that were significantly different between the hemorrhage and nonhemorrhage group on univariate analysis (P < 0.05), or nearing statistical significance (defined as P < 0.1). Probability values <0.05 were defined as statistically significant. For statistical analysis, IBM's SPSS (version 21.0) was used.

Patients' characteristics
Seven hundred and fifty-one patients who had undergone GKRS between December 1993 and December 2016 were selected from the single-center. The patient population consisted of 271 men (36.1%) and 480 women (63.9%) patients with a median age of 38

Discussion
This was the largest study describing the incidence, risk factors and prognosis of pituitary hemorrhage in pituitary adenomas treated with GKRS in a single-center. Clinical pituitary apoplexy is a rare event with a prevalence of 0.6%-10% in previous studies [2,[5][6][7]. In contrast, pituitary hemorrhage including clinical and subclinical pituitary apoplexy, is not an uncommon phenomenon with an incidence of 14%-28% [5,14,25]. Symon et al. [26] reported the incidence of pituitary hemorrhage was 18.1% based on 320 cases of postsurgical specimens. However, this was a cohort with a high proportion of giant or large recurrent pituitary adenomas. Yousem et al. [27] reported a prevalence of 26% of pituitary hemorrhage based on MRI or surgical findings in 68 pituitary adenomas treated medically or surgically, or with radiotherapy. In the study of Liu et al. [14], the incidence of pituitary hemorrhage was 28.1%. In this study, 89.2% of patients were macroadenomas. In the current study, our data showed the incidence of pituitary hemorrhage was 17.0% in patients treated with GKRS totally, which was similar with previous studies [5,14,25]. Eighty-eight patients (11.7%) presented with pituitary hemorrhage before GKRS, which were mainly spontaneous hemorrhage. Fifty-eight patients (7.7%) developed new or worsened pituitary hemorrhage after GKRS. After excluding the 3 patients with new or worsened pituitary hemorrhage due to tumor regrowth, 55 patients (7.3%) were left. This was the first study reporting the incidence of new or worsened pituitary hemorrhage after GKRS. The median time of the 55 patients developed new or worsened pituitary hemorrhage was 18.9 (range 3.1-70.7) months. Pituitary adenomas could present with spontaneous hemorrhage. As the high incidence of new or worsened pituitary hemorrhage after GKRS, GKRS might play an important role in the pathogenesis of pituitary hemorrhage. GKRS might be a precipitating factor of pituitary hemorrhage.
Predisposing and precipitating factors of pituitary hemorrhage including arterial hypertension, intracranial pressure, major surgery, closed head trauma, dynamic tests, angiography, DAs, radiotherapy and anticoagulant therapy have been widely reported [9,15]. However, only a small number of cases with pituitary hemorrhage associated with radiation were reported. Literature reviews on the precipitating factor of radiation for pituitary hemorrhage were showed in Table 3. The incidence, risk factors and prognosis of pituitary hemorrhage in pituitary adenoma patients treated with GKRS have not been reported thus far. In 1977, Weisberg et al [15] reported 14 of 300 patients developed clinical pituitary apoplexy. Among of them, 8 patients with pituitary apoplexy were associated with radiotherapy. Wakai et al. [10] investigated the occurrence of pituitary hemorrhage from 560 cases of pituitary adenomas treated with operation. Apoplectic episodes occurred shortly after the completion of postoperative radiotherapy in 3 cases. One patient occurred hemorrhage with no acute episode after radiation. Zhang et al. [28] investigated the clinical features, diagnosis, treatment and outcomes of subclinical pituitary apoplexy in 185 consecutive patients. Among of them, 5 patients received preoperative radiotherapy. Radiation might be potentially risky in these patients.   The pathophysiology of pituitary hemorrhage is not completely understood. Vascular flux reduction and intra or postoperative hypotension may be proposed mechanisms of pituitary hemorrhage after surgical procedures [29,30]. Pituitary hemorrhage after angiographic procedures may be related either to vasospasm or blood pressure fluctuations [31]. The increased metabolic needs of the tumors induced by dynamic testing may cause pituitary hemorrhage [32,33]. Patients treated with anticoagulation therapy have an unavoidable risk of bleeding [34]. Besides, pituitary hemorrhage can occur spontaneously without any precipitating factor. A rapidly expanding adenomas with poor vascularity that outgrow their blood supply or because ischemia (and thus infarction) occurs following compression of infundibular or superior hypophyseal vessels may result in pituitary hemorrhage [35,36]. Because of the infrequent occurrence, and lacking of case control study, a direct causal relationship between radiation and pituitary hemorrhage has not been confirmed. It is postulated that pituitary hemorrhage may occur following tumor infarction or necrosis after GKRS. Vascular changes after pituitary irradiation often result in chronic hypoperfusion of the pituitary gland and have been associated with both pituitary infarction and hemorrhage [37]. In addition, GKRS might damage the tumor neovascularity to produce hemorrhage [38]. Interestingly, the median time of tumor shrinkage was 14.9 months in this study, which was a little shorter than that of new or worsened pituitary hemorrhage after GKRS. Because of the high incidence and interesting occurrent time, new or worsened pituitary hemorrhage after GKRS might not be a coincidence.
In the current study, pituitary hemorrhage before GKRS was significantly associated with NFPAs and suprasellar extension. New or worsened pituitary hemorrhage after GKRS was related with NFPAs. We also investigated the relationship between gender, age, tumor volume, surgery, parasellar invasion, hypertension, diabetes mellitus, DAs, anticoagulants, angiography and pituitary hemorrhage, but did not find any association. Pituitary apoplexy was observed more frequently in NFPA patients than functioning tumors [39]. Moller-Goede et al. [39] analyzed incidence and risk factors of clinical pituitary apoplexy by reviewing 574 patients with pituitary adenoma. Forty-two patients (7.3%) were diagnosed as clinical pituitary apoplexy. The incidence of clinical pituitary apoplexy in nonfunctioning tumors and functioning tumor were 16.5% and 5.0% respectively. Male and nonfunctioning tumor type were found to be independent risk factors for clinical pituitary apoplexy (p<0.001). Suprasellar extension was associated with pituitary hemorrhage before GKRS. Perhaps, suprasellar extension can compress the infundibular or superior hypophyseal arteries [40], and be risky for pituitary hemorrhage.
In the current study, we did not find relationship between pituitary hemorrhage and tumor regrowth after GKRS. However, every coin has two sides. The course of new or worsened pituitary hemorrhage after GKRS was variable. New or worsened pituitary hemorrhage after GKRS could result in GKRS treatment failure or tumor shrinkage. On the one hand, acute headache, visual abnormalities, hypopituitarism might occur due to tumor enlargement following new or worsened pituitary hemorrhage after GKRS in some cases, and need surgical resection. Of the 58 patients with new or worsened pituitary hemorrhage after GKRS, 5 patients received surgical resection for clinical pituitary apoplexy were defined as GKRS treatment failure. On the other hand, pituitary adenomas could be destroyed by new or worsened pituitary hemorrhage after GKRS. Tumor shrinkage might be accelerated by hemorrhage after GKRS. Therefore, we postulated one of potential mechanisms of GKRS treatment for pituitary adenomas may via radiation-induced pituitary hemorrhage.
Although the current study is the largest describing the incidence, risk factors and prognosis of pituitary hemorrhage in pituitary adenomas treated with GKRS in a single-center, there were several limitations should be noticed. Firstly, it was a single-center retrospective study and there were information and selection biases. Because many patients came from long distance, there were only 741 patients undergoing follow-up evaluation in our hospital. Thus, many patients were lost to follow up. It could cause selection bias in this study. Secondly, the diagnosis of pituitary hemorrhage was mainly based on MRI. Only a small proportion of patients underwent surgery. Besides, some patients did not undergo regular follow-up MRI. Therefore, some micro-bleedings in pituitary adenomas might be missing. Thirdly, this was not a case-control study, the role and mechanism of GKRS on new or worsened pituitary hemorrhage after GKRS needed further study.
To conclude, we have demonstrated the incidence, risk factors and prognosis of pituitary hemorrhage in pituitary adenoma patients treated with GKRS. In this study, 11.7% of patients were diagnosed with pituitary hemorrhage before GKRS, 7.3% of patients developed new or worsened pituitary hemorrhage after GKRS (excluding 3 patients with new or worsened pituitary hemorrhage due to tumor regrowth). The median time to new or worsened pituitary hemorrhage after GKRS was 18.9 (range 3.1-70.7) months. Overall, 17.0% patients were diagnosed with pituitary hemorrhage in the entire series. NFPAs and suprasellar extension were associated with pituitary hemorrhage before GKRS. NFPA was an independent risk factor for new or worsened pituitary hemorrhage after GKRS. New or worsened pituitary hemorrhage after GKRS could result in GKRS treatment failure. GKRS might be a precipitating factor of pituitary hemorrhage.