The importance of preserving the superior hypophyseal artery infundibular branch in craniopharyngioma surgery

Postoperative pituitary dysfunction, a critical problem in the treatment of craniopharyngiomas, can occur even when the pituitary stalk is preserved. We hypothesized that compromise of the primary superior hypophyseal artery (pSHA) might be related to this occurrence. We performed a retrospective review of 131 patients with craniopharyngioma who underwent surgery from April 2009 to September 2021. The inclusion criteria were initial surgery, endoscopic transsphenoidal surgery, preoperative normal pituitary function or pituitary dysfunction in one axis, and morphological preservation of the pituitary stalk. The branches of the pSHA consist mainly of the chiasmatic branches (Cb), infundibular branches (Ib), and descending branches (Db). We analyzed the association between postoperative pituitary function and preservation of these branches. Twenty patients met the criteria. Preoperative anterior pituitary function was normal in 18 patients, and there was isolated growth hormone deficiency in two patients. No patient had preoperative diabetes insipidus (DI). Anterior pituitary function was unchanged postoperatively in eight patients. Of these eight patients, bilateral preservation of pSHA Ib was confirmed in seven patients. Bilateral preservation of pSHA Ib was the only factor associated with preserved anterior pituitary function (p < 0.01). Fifteen patients were free of permanent DI, and the preservation of any given pSHA branch produced no significant difference in the postoperative occurrence of permanent DI. Our study shows that bilateral preservation of pSHA Ib provides favorable postoperative anterior pituitary function in craniopharyngioma surgery; however, such preservation may have little effect on the postoperative occurrence of DI.


Introduction
Craniopharyngioma is a benign tumor derived from embryonic remnants of Rathke's pouch. Postoperative pituitary dysfunction is one of the most important complications of craniopharyngioma because the tumor develops from the pituitary stalk, which plays an important role in endocrine function. Thus, surgeons have attempted to preserve the pituitary gland and pituitary stalk to maintain pituitary function; however, postoperative pituitary dysfunction has often occurred despite these efforts [26,39,41].
The superior hypophyseal artery (SHA) is an important artery feeding the optic nerve, pituitary stalk, and pituitary gland. The SHA branches range from 1 to 6 (mean: 1.83 to 3.2) in number, and the most proximal artery arising from the internal carotid artery (ICA) is referred to as the primary SHA (pSHA) [9,37,38]. One-third of pSHAs originate from the ophthalmic segment of the ICA distal to the distal dural ring (DDR), and the other two-thirds originate from the clinoidal segment of the ICA proximal to the DDR [9,29,30,37,38]. The artery arising from the distal pSHA is referred to as the secondary SHA (sSHA) [9,37]. The pSHA forms three main sets of branches, and the branches are classified as the pSHA chiasmatic branches (pSHA Cb), the pSHA infundibular branches (pSHA Ib), and the pSHA descending branch (pSHA Db) [4,9,37,38].
During craniopharyngioma surgery, in addition to preserving the pituitary stalk, the surgeon must also take care to preserve the pSHA to protect visual function. However, the effect of pSHA preservation on postoperative pituitary function is still unknown. We hypothesized that the vascularity of the pituitary stalk is essential for maintaining postoperative pituitary function. This study describes the preservation of pSHA branches and the associated endocrine outcomes.

Patients
We performed a retrospective review of 131 patients with craniopharyngioma who underwent surgery at our hospital and related hospitals from April 2009 to September 2021. Patients who met the following inclusion criteria were selected: initial surgery, endoscopic transsphenoidal surgery, normal preoperative pituitary function or single-axis pituitary dysfunction, and morphological preservation of the pituitary stalk on intraoperative videos. We excluded patients with recurrent cases, patients treated with nontranssphenoidal surgery, and patients with missing details, thus resulting in 74 included patients. Among intraventricular types of tumors (Kassam type 4) [19], patients in whom the infrachiasmatic corridor was wide underwent endoscopic transsphenoidal surgery [1,32]. Patients with other Kassam type 4 cases underwent craniotomy. Of these 74 patients, we excluded 20 patients who had severe preoperative hypopituitarism (deficiency of two or more axes). We reviewed the videos of the remaining 54 patients and found that the pituitary stalk was sacrificed in 31 (a sacrificed pituitary stalk was defined as a completely disconnected pituitary stalk and pituitary gland). Three patients were excluded because their videos were unavailable (Fig. 1).

Ethics approval
This retrospective study was conducted following institutional review board approval at Nagoya University Hospital (No. 2019-0254).

Preservation of the pSHA
The surgical procedures were performed according to the technique described in our previous reports [16,17,[23][24][25]34]. Two neurosurgeons blinded to the endocrinological outcomes of the patients confirmed the branching of the pSHA in intraoperative videos before and after tumor resection ( Fig. 2A-D). The pSHA branches were defined as follows: The pSHA Cb runs superiorly to supply the optic chiasm. The pSHA Ib runs medially to reach the upper pituitary stalk directly. The pSHA Db runs downward along the pituitary stalk toward the upper surface of the pituitary gland and/or sellar diaphragm. Preservation of branches of the pSHA was defined as the presence of these arteries after tumor resection and was determined by reviewing the surgical field on the patient videos. From 2018 onward, indocyanine green (ICG, 12.5 mg, i.v.) fluorescence imaging was also performed intraoperatively to evaluate blood flow in preserved arteries (Fig. 2E, F).

Evaluation of pituitary function
Anterior pituitary function was evaluated preoperatively and postoperatively by loading tests and fasting hormone levels. The methods used for evaluation of pituitary function are published in our previous report [24]. Briefly, growth hormone deficiency (GHD) was diagnosed on the basis of an observation of low serum levels of insulin-like growth factor 1 (IGF-1) adjusted for age and sex and low response in the growth hormone-releasing hormone (GRH) or growth hormone-releasing peptide 2 (GHRP-2) test, i.e., a peak serum growth hormone (GH) level of < 3 ng/ml in the GRH test or < 9 ng/ml in the GHRP-2 test [3]. Central hypothyroidism was diagnosed when serum thyroxine levels were low in the absence of primary hypothyroidism. Central adrenal insufficiency (CAI) was diagnosed on the basis of low morning serum cortisol levels with concomitant clinical symptoms of hypoadrenalism without primary adrenal insufficiency. Hypogonadotropic hypogonadism was diagnosed in men when serum testosterone levels were low and in premenopausal women with low serum gonadotropin levels when menstrual irregularity was evident. Diabetes insipidus (DI) was defined as a daily urine volume over 3000 ml and a urine specific gravity under 1.005 along with at least one of the following symptoms: excessive thirst or serum sodium over 145 mmol/l [7,10,24]. Postoperative DI was managed by nasal or oral desmopressin administration. Permanent DI was defined as the continuation of these clinical symptoms 3 months later, treated by the administration of desmopressin. Postoperative hormone levels were assessed at 1 week and 3 months after surgery.

Study details
Age, sex, tumor size, and Kassam type were obtained from medical records. The tumor resection rate was evaluated by intraoperative findings and gadolinium contrast-enhanced magnetic resonance imaging (MRI) obtained within 1 week after surgery. Gross total resection (GTR) was defined as resection that left no residual tumor, subtotal resection (STR) was defined as less than 100% but more than 90% removal, and partial resection (PR) was defined as less than 90% removal. Preservation of the pSHA branches was defined as the presence of a relevant artery after tumor removal as determined from intraoperative video. Patients who maintained their pituitary function postoperatively and patients whose pituitary function deteriorated postoperatively were investigated for bilateral preservation, unilateral preservation, or bilateral sacrifice of each pSHA branch.

Statistical analysis
Statistical analysis was performed by the chi-square test and Student's t test. A p value less than 0.05 was considered statistically significant. The statistical analysis software employed was IBM SPSS Statistics version 28 (IBM, Armonk, NY, USA).

Results
Of the 131 patients screened for eligibility, this study included eleven male and nine female patients. The mean age of the participants at the time of surgery was 41.6 (4-70) years. The mean maximum tumor diameter was 28.1 (13-52) mm. Kassam type 1 was diagnosed in nine patients (45%), type 2 in six patients (30%), type 3 in three patients (15%), and type 4 in two patients (10%). Preoperative anterior pituitary function was normal in 18 patients (90%), and there was GTR was achieved in 18 patients (90%), and STR was achieved in two patients (10%). Tumors recurred in three patients after GTR and in one patient after STR. The total tumor recurrence rate was 20%. All four patients who experienced tumor recurrence underwent a second operation. Among these four patients, GTR with sacrificing the pituitary stalk was achieved in two patients; STR with preservation of the pituitary stalk was achieved in two patients who were kept under observation and needed no further treatment because there was no tumor regrowth. None of the patients in this study underwent radiotherapy after surgery. Regarding pathological diagnoses, one tumor belonged to the squamous papillary type, and 19 belonged to adamantinomatous types.  Table 1).
The bilateral pSHA Ib preservation rate was significantly higher in patients who retained full anterior pituitary function postoperatively (the maintenance group) than in those whose function deteriorated (the deterioration group) (87.5% vs. 25%, p < 0.01). On the other hand, there was no significant difference in the bilateral preservation rate of the pSHA Cb or the pSHA Db between the maintenance group and the deterioration group (100% vs. 75%, p = 0.13; 37.5% vs. 8.3%, p = 0.11). There was also no significant difference between the two groups regarding age, sex, maximum tumor diameter, Kassam type, or tumor resection rate ( Table 2). The preservation of any given branch of the pSHA was not associated with postoperative permanent DI. Likewise, there were no significant differences in other factors (Table 3).

Discussion
In surgery for craniopharyngiomas, preservation of the pituitary stalk is essential for maintaining pituitary function [14,31]. Previous studies have shown that the transinfundibular type (Kassam type 2) has poor postoperative endocrinological outcomes [8,22,35]. In another study, researchers reported that firm attachment between the tumor and hypothalamus, severe hypothalamus deformation on preoperative MRI, and GTR were associated with poor endocrinological outcomes [42]. However, to the best of our knowledge, no study has been conducted in which researchers investigate the association between the vascularity of the pituitary stalk and endocrinological outcomes.
pSHA Ib forms a pituitary portal system after penetrating the median eminence zones. Axons of parvocellular neurons arising from the paraventricular hypothalamic nucleus and the arcuate nucleus terminate in the median eminence and release hypothalamic hormones into the pituitary portal system (Fig. 3A) [11,27]. Therefore, preservation of pSHA Ib is important to maintain anterior pituitary function. Indeed, our results suggest the importance of bilateral preservation of pSHA Ib for endocrinological outcomes.
The pSHA Db courses to the anterior pituitary gland and/ or diaphragm, but bilateral preservation of the pSHA Db had no association with favorable postoperative anterior pituitary function in this study. The reason for this finding is that the blood supply of the pituitary gland is complex. Previous reports have shown that the pSHA Db anastomoses with the contralateral pSHA Db, the pSHA Ib, and the inferior hypophyseal artery (IHA) [4,6,21,33,37,40]. Therefore, it has been reported that the pSHA Db can be sacrificed if it impedes tumor resection [4,37], and our results also support this suggestion. Bilateral preservation of the pSHA Cb was not associated with maintaining pituitary function because this branch supplies the optic chiasm [9,13,15,29,30,37,38].
Despite bilateral preservation of pSHA Ib, three patients showed postoperative deterioration of anterior pituitary function. One patient had a Kassam type 4 tumor, and the hypothalamus was injured in this patient. The other two patients exhibited firm attachment between the tumor and the pituitary stalk, and the pituitary stalk was severely injured by the manipulation of the tumor for resection in these patients. Therefore, morphological preservation and  Four types of anatomic variations of the pSHA have been reported: the candelabra-like type, in which the pSHA splits off to form three major branches at the same point (pSHA Cb, pSHA Ib, pSHA Db); the tree-like or early optic branching type, in which the branches split off one by one; the nodescending type, in which the pSHA Db is not present; and the no-optic type, in which the pSHA Cb is not present [9,12,20,37]. The hemodynamics of pSHA Ib are lateralized, and this vessel may branch from the unilateral ICA. Moreover, the branches of the SHA and the perforators originating from the posterior communicating artery anastomose around the pituitary stalk, forming circuminfundibular anastomosis [12,36,37]. The anastomosis may play a role in maintaining pituitary function in patients with unilateral preservation of pSHA Ib. Indeed, postoperative pituitary function was preserved in some patients with unilaterally preserved pSHA Ib in this study.
On the other hand, the present study showed that preserving any branches of the pSHA had little influence on postoperative posterior pituitary function. This may be because the pSHA mainly feeds the anterior part of the pituitary stalk and anterior pituitary gland. The posterior part of the pituitary stalk consists of axons extending from the paraventricular hypothalamic nuclei and supraoptic nuclei to the posterior pituitary lobe. The posterior part of the pituitary stalk is supplied by the sSHA [4,9,18,20,37,38], and the posterior pituitary lobe is supplied by the IHA and other collateral flow (Fig. 3) [4,28]. This may be the reason that there was no association between preservation of the pSHA and favorable postoperative posterior pituitary function in this study.
The preservation of the pituitary stalk during surgery is controversial, as it is known to increase the risk of tumor recurrence [2,5]. Our concept for craniopharyngioma surgery is to preserve postoperative pituitary function. We avoided aggressive resection if the tumor had a firm attachment on the pituitary stalk and pSHA Ib. If the tumor recurs, reoperation is the first choice. At that time, if the pituitary function is poor, we consider GTR with sacrificing the pituitary stalk. Radiotherapy is performed for unresectable lesions, for example in cases of hypothalamic infiltration. We suggest performing GTR while sacrificing the pituitary stalk when sacrificing pSHA Ib cannot be avoided and the morphology of the pituitary stalk is extremely thin.  This study has several limitations. First, the sample size is relatively small. Therefore, we could not compare unilateral preservation in the pSHA Ib group with bilateral sacrifice in the pSHA Ib group in detail. For the same reason, we could not perform multivariate analysis; thus, we will need to increase the number of cases to examine these issues. Second, we selected only patients in whom the pituitary stalk was successfully preserved, which could lead to selection bias. Therefore, there might have been no significant difference between the Kassam type and postoperative endocrinological outcome.
In the future, the flow volume of pSHA Ib after tumor resection should be evaluated using ICG fluorescence imaging. Overall, when there is a chance of preserving the pituitary stalk, surgeons should apply maximum effort to preserve pSHA Ib, the artery that feeds the pituitary stalk.

Conclusions
Our study suggests that bilateral preservation of pSHA Ib is favorable for postoperative anterior pituitary function outcomes in craniopharyngioma surgery. In contrast, bilateral preservation of the pSHA Ib may not affect postoperative posterior pituitary function. For patients with a chance of preserving the pituitary stalk, favorable postoperative anterior pituitary function is more likely if pSHA Ib is preserved bilaterally. Therefore, pSHA Ib should be preserved during surgery when possible.

Declarations
Ethics approval and consent to participate This retrospective study was conducted following institutional review board approval at Nagoya University Hospital (No. 2019-0254). All the procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional or national research committee and with either the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. This study obtained informed consent from all the patients.

Conflict of interest
The authors declare no competing interests.