Diaphragma Sellotomy: A Safe Technique to Confirm Adequate Decompression of Optic Chiasm

Abstract

Optic chiasm decompression for preservation of vision is often the primary surgical goal for patients with pituitary tumors. Descent of the diaphragma sellae (DS) is an intraoperative surrogate marker of adequate chiasm decompression. DS may not always descend in an obvious or symmetrical manner, leaving uncertainty to whether the operation was successful. We propose a technique of intentionally incising the DS to ensure adequate chiasm decompression. Here we present patients with pituitary tumors who underwent transsphenoidal surgery and DS incision when the DS was not easily identified and/or did not descend. The approximately 3-mm incision under endoscopic guidance allowed for direct visualization of the suprasellar cistern and optic chiasm. Cerebrospinal fluid (CSF) leak was repaired using a nasoseptal flap in 4 cases, while intradural substitute and thrombin glue were utilized in another case. Five patients with pituitary macroadenomas (average size: 6.4 cm³) had endoscopic endonasal transsphenoidal resection. Vision improved in all cases postoperatively. There were no postoperative complications or CSF leaks at 1 year. Diaphragma sellotomy ensures chiasm decompression with minimal risk to the patient with current reconstructive techniques and without the need for intraoperative magnetic resonance imaging (iMRI) and reduction of the need for repeated surgical intervention.

1. Introduction

Descent of the diaphragma sellae (DS) during endonasal endoscopic surgery is what surgeons use as a surrogate to determine whether or not enough tumor has been removed to decompress the optic chiasm [1, 2]. The optic chiasm is not directly visible even after removal of the tumor [3]. Ideally, once the tumor is removed, the DS, which is a barrier between the sella and the subarachnoid space, would bow into the field, suggesting successful decompression of the optic chiasm [1, 2]. An ongoing issue in endoscopic transsphenoidal surgery is that the DS is not always identifiable as often times, and the tumor develops a capsule that adheres to the DS, which makes it difficult for the surgeon to differentiate between the two [1, 2, 3]. Even when utilizing an endoscope, which allows direct visualization of the sella, multiple intraoperative MRI imaging studies have revealed that there is approximately a 30–66% chance that one may think that they have decompressed the optic chiasm while they have not actually done so [4, 5, 6, 7, 8, 9]. Highly successful outcomes associated with utilization of modern techniques of intraoperative cerebrospinal fluid (CSF) leak repair [10, 11, 12, 13, 14, 15, 16] allow surgeons to develop innovative approaches to tumor resection, which, in our retrospective study, may ensure decompression of the optic chiasm and reduce the chance of reoperation, without the use of costly intraoperative imaging. In this study, we analyzed the safety and feasibility of a novel technique of “diaphragma sellotomy” to assess, with direct visualization, optic chiasm decompression and extent of tumor removal, without the use of intraoperative imaging. We present a series of five patients with pituitary macroadenomas (average size: 6.4 cm³) who have undergone endoscopic endonasal transsphenoidal resection. The goal of surgery in each case was chiasm decompression, as each patient suffered rapid and severe vision loss. Successful intraoperative diaphragma sellotomy confirmed chiasm decompression and subsequent CSF leak repair resulted in no postoperative complications. The optic chiasm was directly visualized in all cases. Vision improved in four out of five cases postoperatively, an exception occurred in a patient with a recurrent Rathke’s cleft cyst that required multiple resections at other institutions. There were no postoperative CSF leaks at 1 year. This novel technique is a facile and safe way to ensure the goals of surgery have been met with minimal risk.

2. Methods

A retrospective chart review was conducted that analyzed outcomes in patients with pituitary adenomas associated with profound visual deterioration who underwent transsphenoidal surgery and diaphragma sellotomy. The diaphragma sellotomy procedure was performed only when the DS was not easily identified and/or did not descend symmetrically into the field of view. The 3-mm incision allowed for direct visualization of the suprasellar cistern and optic chiasm, which allowed for confirmation of optic chiasm decompression and avoided the need for re-operation. The resultant CSF leaks were repaired using a nasoseptal flap in four cases, and intradural substitute and thrombin glue were utilized in another case. All patients had MRIs and neurological examination of visual fields prior to surgery and at 3 months post-operatively. All patients were neurologically assessed in clinic one-year post surgery.

3. Results

3.1 Demographic and presenting features

After a retrospective chart review of patients undergoing pituitary surgery at our institution, five were selected as having additionally undergone diaphragma sellotomy to assess, with direct visualization, optic chiasm decompression, and extent of tumor removal. Two patients were female and three were male. Their average age at presentation was 48.6 (range 18–69 years). Average body mass index (BMI) was 29.02 kg/m² (range 24.7–35.4 kg/m²). Average size of pituitary macroadenomas was 6.4 cm³ (range 2.3–13.2 cm³). All patients initially presented with peripheral field vision loss, which improved in all cases but one, postoperatively. The patient in whom vision did not improve post-operatively had a history of a recurrent Rathke’s cleft cyst that required multiple resections at other institutions. There were no postoperative complications or CSF leaks at a 1-year follow-up clinic visit.

3.2 Case illustration

A 69-year-old male with past medical history significant for diabetes mellitus type II presented to our institution’s neurosurgery clinic with chief complaints of frontal headaches and peripheral vision loss. The patient first noticed changes in his vision approximately one and half years prior to his visit. He was previously diagnosed with a pituitary tumor via magnetic resonance imaging (MRI) at a different institution. However, his vision had progressively worsened over a 3-month period of time, thereby taking a toll on his performance as an avid pickleball player, and eventually led to his presentation at our facility. Neurological exam was unremarkable except upon the assessment of visual fields, and it was determined that the patient had marked left temporal and superior right quadrant visual field deficits. Pre-operative MRI revealed a heterogeneously enhancing sellar/suprasellar mass, which was 2.2 cm anteroposterior x 2.5 cm transverse x 3.2 cm craniocaudal (Figure 1) and was therefore determined to be the root cause of significant compression of the patient’s optic chiasm. Intraoperatively, the tumor was removed off of the diaphragma sellae and pseudocapsule. Because the diaphragma sellae did not herniate and was not decompressed fully, tuberculum sellae was then further opened using Kerrison rongeurs, and the asymmetrically descended diaphragma sellae was opened from above, superiorly, using the retractable blade in a horizontal fashion (Figure 2). Once the supradiaphragmatic space was identified, CSF and the optic chiasm were visualized, and it was confirmed that the optic chiasm had been decompressed (Figure 3). After thorough irrigation and hemostasis were achieved, an intradural graft along with a bone graft and a nasoseptal flap were cut and fashioned to be placed in order to cover the defect (Figure 3). Surgicel™, Duraseal™, and Gelfoam™ were then placed on top to finalize the closure and observe for a possible CSF leak, the latter of which never occurred. Immediately after surgery the patient noticed improvement in his left visual field. A 3-month post-operative MRI demonstrated near total resection of the pituitary tumor along with a fully decompressed optic chiasm (Figure 4). At the 3-month post-operative clinic visit, the patient stated that his vision continued to improve, and he was eager to share that he had finally resumed playing pickleball. Upon neurological exam, the patient’s visual fields were fully intact.

Figure 1.

Pre-operative MRI. A pre-contrast T1 sagittal demonstrating a sellar/suprasellar mass (white arrowhead) isointense to gray matter, compressing the optic chiasm (white arrow). b pre-contrast T1 coronal demonstrating the compressed optic chiasm (white arrow) and mass (white arrowhead).

Figure 2.

Intraoperative images of asymmetric descent of DS and incision. A asymmetric descent of DS (white arrow). b residual pituitary gland (white arrow). c incision area (white arrow) in DS. d incision made (white arrow) in DS.

Figure 3.

Intraoperative images of confirmation of decompression and closure. A optic chiasm (white arrow). b Sella (white arrow), diaphragma (white arrowhead), suprasellar space (black arrowhead). c Intradural graft (white arrow). d bone graft across the Sella (white arrow). E Nasoseptal flap (white arrow) covering the defect, no cerebrospinal fluid leak observed.

Figure 4.

3-month post-operative MRI. A pre-contrast T1 sagittal demonstrating near total resection of tumor, with optic nerve (white arrow), pituitary stalk (white arrowhead) and fat (black arrow). b pre-contrast T1 coronal demonstrating the decompressed optic chiasm (white arrow) and fat (black arrow).

4. Discussion

Records of pituitary tumor surgery go back to the early 1900s when common surgical techniques included palliative craniectomies, subfrontal, or subtemporal craniotomies, as well as transnasal transethmoidal, sublabial endonasal-transseptal, and transsphenoidal methods. Unfortunately, mortality rate ranged from 66 to 21%, which was associated with palliative craniectomy and transsphenoidal approach, respectively [17]. However, between 1910 and 1925, Cushing focused exclusively on the transsphenoidal approach which he utilized in 231 pituitary tumor cases and was able to achieve a mortality rate of only 5.6%. It is indubitable that the advent of endoscope allowed for greater visualization of neuroanatomical structures, especially during transsphenoidal surgery, and in addition to modern understanding of microbiology, as well as utilization of aseptic surgical technique, the surgical mortality rate for this procedure is now approximately 0.6% [18].

The decompression and preservation or restoration of vision is often the primary surgical goal in the treatment of patients harboring pituitary macroadenomas with severe optic chiasm compression with up to 80.9% of patients experiencing post-operative improvement in their vision [19]. Despite the success of transsphenoidal surgery overall at improving vision, this success is still dependent on several factors that can be improved upon, including surgical technique. Descent of the diaphragma sellae (DS), a thin, tenuous structure that forms the roof of the sella turcica and covers the pituitary gland, during endonasal endoscopic transsphenoidal surgery is the surrogate marker of adequate chiasm decompression [1, 2, 3, 20]. In management of large pituitary lesions, this structure is often difficult to identify, or may be violated by the tumor. At times, it may not descend symmetrically, leaving the surgeon unsure whether the goals of surgery have been met. Due to the difficulty of distinguishing normal pituitary gland, tumor, tumor capsule, and the diaphragma sella, intraoperative MRI (iMRI) has become an increasingly utilized, though cost-prohibitive tool in pituitary surgery, without which the surgeon is able to predict extent of resection in only 65% of cases [21]. Furthermore, the surgeon is often left to finish surgery questioning whether the optic chiasm is decompressed and hoping for future spontaneous diaphragma descent and chiasm decompression if adequate tumor debulking was performed. Our proposed technique of intentionally incising the presumed diaphragma sella allows for assurance of adequate chiasm decompression at the time of surgery. Intraoperatively, the presumed diaphragma sella can be incised to visualize into the subarachnoid space and suprasellar cistern and confirm the extent of optic chiasm decompression. The small low flow CSF leak can be repaired with a number of techniques at the disposal of the skull base surgeon. It is common neurosurgical practice to utilize abdominal fat grafting when a large cavity is encountered after tumor resection to fill the “dead space,” as well as a Durepair™ inlay, Duraseal™, and Surgicel™ [22, 23, 24]. Nasoseptal flaps are typically reserved only for large, high flow CSF leaks [25]. Current trends in pituitary surgery advocate against intentional opening of the diaphragma sella, as this increases the risk of a postoperative CSF leak. More recent studies have shown that despite intraoperative CSF leak, with proper reconstruction techniques, postoperative CSF leak rates are exceedingly low and range from 0.6 to 2.3% and can be easily managed or even completely prevented with utilization of lumbar drainage [11, 24, 26]. With the main goal of vision preservation and restoration, diaphragma sellotomy ensures the goals of surgery are met with minimal risk to the patient, while also allowing for pituitary gland preservation; repeat surgery stemming from patients reporting marginal clinical improvement along with radiographic studies demonstrating persistent optic chiasm compression, can be avoided.

5. Conclusion

Optic chiasm decompression is often the primary surgical goal for patients with pituitary macroadenomas. Descent of the diaphragma sellae (DS) indicates chiasm decompression. DS is not always identifiable and its descent not always symmetric. Intraoperative MRI (iMRI) has become an increasingly utilized, though cost-prohibitive tool to confirm decompression. Diaphragma sellotomy ensures chiasm decompression with minimal risk to the patient with current reconstructive techniques and without the need for iMRI. Repeat immediate or delayed surgery to ensure chiasm decompression can be avoided. Diaphragma sellotomy is a facile and safe way to ensure the goals of surgery have been met with minimal risk. Utilization of this novel technique may allow for more rapid recovery of vision and prevention of repeated surgical intervention.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflicts of interest.

Informed consent

Informed consent was not required from participants included in the study. All of the procedural outcomes and test results analyzed were previously collected as part of the course of routine treatment, and no additional interventions were required, thereby waiving the need for informed consent. The authors’ institutional review board (IRB) approved waiver of informed consent/authorization for the study.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.