There is a growing recognition of an association between pituitary adenoma and existence of intracranial aneurysms which increase the risk of SAH. Reports in the literature suggest the association lies somewhere between 1.4 % (2/144 over 4 years, Acqui et al. 1987) and 7.4 % (7/95 over 5 years, Wakai et al. 1979). In a series of 800 patients with pituitary adenoma, the incidence of intracranial aneurysm was 2.3 % (Oh et al. 2012). Intracranial aneurysms are probably due to microanatomical changes in the cerebral circulation from compression or traction with consequent hyperaemia (Pia et al. 1972; Goyal et al. 2012). Hori et al. (1982) suggested that an internal carotid artery (ICA) aneurysm arose and enlarged as a result of a traction mechanism that a GH-secreting macroadenoma exerted as it reduced in size. Oh et al. (2012) reported that that age (p = 0.04) and cavernous sinus invasion (p < 0.001) were independently associated factors, whereas hormone type, immunohistochemistry staining and patient sex failed to demonstrate any statistically significant correlation. Pant et al. (1997) also reported that the association between intracranial aneurysms and pituitary adenoma was a function of increasing age (p < 0.001) but that no independent association existed with hormone secretion or the size and malignancy of tumour. Manara et al. (2011) also claimed no independent association between aneurysm formation and either adenoma size or invasiveness. Interestingly, intracranial aneurysms are also significantly more prevalent in patients with pituitary adenomas when compared to other brain tumours (1.1 %, p < 0.001, Wakai et al. 1979).

A meta-analysis by Vlak et al. (2011), which included 68 studies on 83 populations and 1,450 intracranial aneurysms in 94,912 patients from 21 countries (years 1998–2011), reported the prevalence ratio of unruptured intracranial aneurysms for pituitary adenomas was 2.0 (95 % CI 0.9–4.6). However, compared with patients who did not have the relevant comorbidity or risk factor, sex-adjusted and age-adjusted prevalence ratios were not significantly higher for patients with pituitary adenoma (from a total of 26 studies including 31 populations and 6,000 unruptured intracranial aneurysms). Nevertheless, Oh et al. (2012) performed an age-matched comparison of the prevalence of intracranial aneurysms in 800 patients with pituitary adenoma, which demonstrated an increased prevalence when compared with controls (p = 0.014) with the sixth decade showing the most significant difference (p = 0.039). Patients with acromegaly presented with a significantly higher prevalence of intracranial aneurysms and, presumably, therefore an increased risk of SAH compared with controls (26/150 vs. 5/71). This difference persisted when acromegalic patients were compared with positive controls above the age of 40 years (25/126 vs. 5/52; Manara et al. 2011).

The association between pituitary adenomatous haemorrhagic apoplexy and SAH is rare (Mohanty et al. 1979; Laidlaw et al. 2003; Nakahara et al. 2006; Bao et al. 2007; Kim and Cho 2007) in spite of adenomas being very vascular (Jefferson 1940; Beard et al. 1985). Reports of pituitary haemorrhage in association with SAH simulating aneurysm rupture are also rare (Gazioğlu et al. 2002). The most extensive adenomas are malignant such as those which burst through the tumour capsule and not only form an intracranial mass but also invade the cavernous sinuses and through the arachnoid membrane (Jefferson 1940; Gazioğlu et al. 2002). Extravasation into subarachnoid space may also occur in the absence of suprasellar extension of the tumour (Cardoso and Peterson 1984). Necrotic haemorrhage within pituitary adenoma is the modal cause of apoplexy which can lead to frank SAH. Although rare, it is an important cause of angiogram-negative SAH (Laidlaw et al. 2003).

SAH can follow discharge of necrotic tissue and blood extravasation from the sella turcica either through the diaphragmatic aperture or after its rupture (Wakai et al. 1981) into the adjacent subarachnoid space of basal cisterns. It occurs when the pressure gradient within the sella exceeds the peripheral resistance of adjacent structures as the intrasellar pressure constantly rises proportionally with the extent of intratumoral haemorrhage and localised inflammation in pituitary apoplexy (Itoyama et al. 1990; Gaini et al. 2004; Satyarthee and Mahapatra 2005; Bao et al. 2007). Itoyama et al. (1990) alternatively hypothesised that apoplexy-associated SAH follows the penetration of atypical lymphocytes via ameboid reaction through the tumour capsule.

Causes of extrasellar haemorrhagic extension depend on the intrinsic growth urge of the adenoma and its potential for malignancy, the integrity of the interclinoid ligaments of the sellar diaphragm, and fixation of the optic chiasm (Jefferson 1940). The latter is important as an abnormally fixed optic chiasm can delay onset of visual disturbance on compression by a macroadenoma. Early vasospasm which could lead to SAH may result from secretion of potent vasoactive agents from the pituitary adenoma (Pozzati et al. 1987; Itoyama et al. 1990). Although vasospasm may be attributable to SAH (Itoyama et al. 1990), SAH can subsequently cause cerebral infarction due to further vasospasm in surrounding vessels because of the irritant effect of blood (Bhansali et al. 2005). Refer to cerebral ischaemia in pituitary apoplexy (Chap.​ 9) for more information. Alternatively, rupture of an intracranial aneurysm may mimic or even result in pituitary apoplexy (Shahlaie et al. 2006). Refer to mimicking conditions (Part VI) for other differentials.

In a study by Pant et al. (1997) of 467 patients with pituitary adenomas, 5.4 % (n = 25) were found to have intracranial aneurysms as diagnosed by magnetic resonance imaging (MRI) and cerebral angiography. 97 % of these were found to have intracranial aneurysms within the anterior circulation, 92 % were incidental findings, 12 % had multiple aneurysms, and only 8 % presented with aneurysmal rupture. The association was most commonly observed among patients with nonfunctioning adenomas (8.8 %) and least frequent among those with a prolactinoma (2.4 %) which also reflected the influence of the age factor (Pant et al. 1997). Although the population distribution of intracranial aneurysms was not significantly different according to Oh et al. (2012), Jakubowski and Kendall (1978) report two-thirds of patients with acromegaly involved the cavernous carotid artery, whereas one-third of the aneurysms in the patients with chromophobe adenoma were thus located. Two patients in their study were found with cavernous aneurysms after treatment with local yttrium implantation for acromegaly.

Manara et al. (2011) reported that 67.5 % (27/40) of newly diagnosed aneurysms among patients with acromegaly (26/151) were located in the intracranial tract of the internal carotid artery (vs. 23–42 % in general population). 22.5 % were in the intracavernous segment compared to 2–9 % in the general population, 17.5 % (7/40) were detected at the level of the middle cerebral artery (MCA; vs. 30–42 % in general population), whereas 15 % (6/40) were found in the ACA (vs. 24 % in general population), but none within the vertebrobasilar circulation (vs. 10 % in general population).

Pituitary adenoma-related intracranial aneurysms are hypothesised to occur due to mechanical compression (or invasion) of the regional arteries by the tumour, vasculopathy, microcirculatory, and haemodynamic changes in the regional vessels near the adenoma, skull base deformities, endocrine dysfunction that potentially induces fibromuscular dysplasia of arterial walls (e.g. multiple endocrine neoplasia syndromes) and specific hormonal factors associated with GH-secreting adenomas (usually macroadenomas) having a predisposing influence on aneurysmal formation in up to 50 % secondary to vascular disease (Jakubowski and Kendall 1978; Cardoso and Peterson 1984; Acqui et al. 1987; Weir 1992; Adachi et al. 1993; Bulsara et al. 2007; Manara et al. 2011). Furthermore, the fact that larger aneurysms are often adjacent to pituitary tumours suggests a direct, local effect of growth hormone in addition to generalised arteriectasia (Weir 1992). Additionally, those with multiple endocrine neoplasia syndrome type 1 (MEN-1) are likely associated with Marfanoid pathomorphology or von Recklinghausen’s disease which includes vascular abnormalities and medial necrosis of arterial walls which could contribute to cerebral aneurysm formation (Adachi et al. 1993). Wakai et al. (1979) reported the rate of coincidental intracranial aneurysms was 12 % in 25 cases of acromegaly, 8 % in 11 cases of prolactinomas and 3.4 % in their cases with nonfunctioning chromophobe adenomas. Furthermore, Jakubowski and Kendall (1978) calculated an incidence of 13.8 % for intracranial aneurysms associated with GH-secreting adenomas (4/29), 5.1 % with chromophobe adenomas (6/117) and no cases for basophilic adenomas (0/4). Bilateral aneurysms with pituitary adenomas have also been reported (Weir 1992).