Microsurgical management and long-term outcomes of pineal region tumors: A retrospective study

INTRODUCTION

Pineal region tumors constitute 0.4–1% of all intracranial tumors in adults and 2–6% in children.^[1] Due to the rarity as well as anatomical complexity of the pineal gland Owing to its intricate relationship with the brainstem, posterior third ventricle, and Galenic system of veins, pineal region tumors are challenging surgical lesions.^[2-4] Improvement in the operating microscope with excellent visualization and refinement in microsurgical techniques have considerably improved the outcome of pineal region surgery.^[2] With the further addition of adjuvant therapies, favorable survival is seen even in high-grade lesions.^[5]

We here have presented our experience of operating pineal region tumors over the past 6 years and also have enumerated the post-operative long-term outcome and survival profile.

MATERIALS AND METHODS

This is an observational retrospective study. Ethical approval was taken from the Institutional Review Board. Data were collected for a total of 21 patients with pineal region tumors who were operated on between 2016 and 2023. Both adults and children were included in the study (patients in whom pre-operative blood and cerebrospinal fluid (CSF) germ cell markers were suggestive of germinoma were not operated). All pre-operative and post-operative radiological studies were obtained from the departmental database. Histopathological analyses were done for all operative specimens and final impressions were collected. Pre-operative patient parameters which were considered are symptomatology, Glasgow coma scale (GCS) score and requirement for pre-operative CSF diversion (external ventricular drains (EVD) or ventriculoperitoneal (VP) shunt). Post-operative parameters considered are – postoperative complications, mortality, requirement for postoperative CSF diversion, and length of hospital stay. Outcomes were assessed in the immediate post-operative period as well as over 1–5 years with a median follow-up of 2 years including the requirement for adjuvant therapy based on histopathology report.

Surgical technique

All patients except two were operated on by infratentorial supracerebellar approach (Stein’s approach)^[2-5] in the sitting/semi-sitting position. Pre-operative magnetic resonance imaging (MRI) was studied to delineate the relation between the Galenic system of veins to the tumor [Figure 1a]. Pre-operative MRI and computed tomography (CT) whenever available were used to approximately estimate the tentorial slope to ensure the feasibility of Stein’s approach. Appropriate precautionary measures were taken by the anesthesia team at the time of induction including central venous line, arterial line, and EtCO₂ probe placement. Trans-esophageal echocardiography whenever available was kept as backup. After induction, the operating table was adjusted to give the sitting/semi-sitting position without compromising the airway and making the tent as parallel to the floor as possible [Figure 1b].

Close

Figure 1:

(a) Case 1 - Pre-operative magnetic resonance imaging images. (b) Case 1 - Sitting position photo.

Export to PPT

A midline skin incision is taken and muscles are subperiosteally dissected to expose suboccipital bone. A suboccipital craniotomy is performed until the transverse sinus is just visualized. The dura is opened based on the sinus. Bridging veins are preserved as much as possible and sacrificed whenever needed to reach the quadrigeminal cistern and pineal region. On careful arachnoid dissection and exposing the tumor, a biopsy is taken and sent for the frozen section. Using suction and microinstruments, tumor was gradually debulked and gross total excision (GTE)/near-total excision was achieved. Following absolute hemostasis, watertight duraplasty was done either primarily or using autologous tensor fascia lata/artificial graft (Duragen). The wound was closed in layers. No drain was used.

RESULTS

A total of 21 patients were operated on from 2016 to 2023 in our institution. Among them 10 were male, and 11 were female (three children included). The age range was (8–64) years with a median age of 40 years. Presenting symptoms were headache (100%), vomiting (90%), and giddiness (85%). Pre-operative imaging (CT/MRI) showed hydrocephalus in 17 patients (81%) [Tables 1 and 2]. Tumor size in imaging ranged from (1 to 3.5) cm with a median tumor size of approximately 1.8 cm.

Two patients presented with poor GCS and emergency CSF diversion in the form of EVD was performed in them before surgery [Figure 2] and EVD was removed once their neurological condition improved after surgery [Figure 3]. Two more patients underwent pre-operative VP shunt [Table 3].

Close

Figure 2:

(a) Case 2 – Pre-operative magnetic resonance imaging. (b) Case 2 - Post-operative computed tomography scan.

Export to PPT

Close

Figure 3:

(a) Case 7 – Pre-operative magnetic resonance imaging, (b) Case 7 – Post-operative computed tomography.

Export to PPT

Pre-operative germ cell markers in serum/CSF were negative in 19 patients and inconclusive in two patients for whom the decision for surgery was taken due to the size and mass effect of the tumor. Pre-operative 2D echocardiography was done in all patients to rule out a patent foramen ovale (PFO). None of the patients had a PFO.

Nineteen patients were operated on in the sitting position, one patient in the prone position (due to unfavorable anatomy of tumor for sitting approach), and one in the supine by anterior interhemispheric approach (due to tumor projecting toward anterior third ventricle in MRI). Among the 19 patients, an intraoperative air embolism happened in one patient who could be managed successfully and surgery was completed. In one patient (in the sitting position), venous injury (possibly vein of Galen) with significant bleeding happened after dural opening and attempt to tumor dissection and had to be abandoned. GTE/near total excision was achieved in 20 patients. We did not use any intraoperative neuro-monitoring adjunct as it was not mandated for the cases that we operated.

Post-operative complications

The patient who sustained intraoperative venous injury succumbed in the immediate post-operative period. A post-operative CT scan was obtained in the rest of the patients. One patient was reexplored on payable on death (POD) 1 due to tumor bed hematoma with mass effect but eventually succumbed to death. Another patient expired on POD 6 from septicemia and multi-organ failure. Two patients had persistent hydrocephalus in post-operative imaging with attendant symptoms and, hence, were subjected to VP shunt [Figure 4]. An intraoperative Ommaya reservoir was placed in one patient. One patient had significant post-operative cognitive and sleep cycle disturbance and a long hospital stay but recovered subsequently and was discharged. Another patient had transient hemiparesis and mutism which improved over 7–8 days.

Close

Figure 4:

(a) Case 3 – Pre-operative magnetic resonance imaging. (b) Case 3 – Post-operative computed tomography scan.

Export to PPT

Histopathology

There was a varied spectrum of histopathological results with pineocytoma (6), meningioma (3), pineal region glioma (2), epidermoid cyst (3) [Figure 5], pineoblastoma (1), primitive embryonal tumor (2), mesenchymal tumor (1), and pineal parenchymal tumors of intermediate differentiation (PPTID) (1), papillary tumor of pineal region (1) [Table 4] and [Figures 5-7]. Among meningioma patients, 1 was grade 2 and all gliomas were grade 2 or 3. One patient with pineoblastoma and one with PPTID expired in the post-operative period.

Table 4: Histopathological Spectrum.

Types	Number of patients	Percentage
Pineocytoma	6/20	30%
Pineal region glioma	2/20	10%
Pineoblastoma	1/20	5%
Meningioma	3/20	15%
Epidermoid	3/20	15%
Others	5/20	25%

Close

Figure 5:

(a) Case 6 – Pre-operative magnetic resonance imaging. (b) Case 6 - Intraoperative image. (c) Case 6 – Post-operative computed tomography.

Export to PPT

Close

Figure 6:

(a) Case 5 – Pre-operative magnetic resonance imaging. (b) Case 5 - Pineal mesenchymal tumour showing Mesenchymal pattern( Red circle). H and E stain, 40X Magnification.

Export to PPT

Close

Figure 7:

(a) Case 4 - Pre-operative magnetic resonance imaging (MRI) axial sequence. (b) Case 4 – Pre-operative MRI sagittal sequence, and (c) Case 4 Pineocytoma showing Round sheets of cells (Red oval). 100x magnification.

Export to PPT

DISCUSSION

The sitting position was introduced by Krause which was later refined by Stein.^[6] However, despite considerable advances in microsurgical techniques and adjunct therapy modalities, the management of pineal region tumors remains challenging.^[2,4,5,7] First, these are rare tumors. Second, the anatomical location of the pineal gland is close to the tectal plate and the Galenic system of veins which make surgical dissection tricky. Third, the varied pathological spectrum of these tumors complicates the decision-making even more.^[5,7]

Open microsurgical excision of pineal region tumors offers various advantages compared to lesser invasive modalities such as endoscopic third ventriculostomy (ETV) or stereotactic procedures as it allows good visualization of the pathology, greater excision of the mass with better tissue yield for histopathology, reduction of mass effect and alleviation of hydrocephalus by opening up CSF pathways and lesser chance of injury to surrounding critical vascular structures.^[3,5]

For suspected germinomas, if initial blood and CSF germ cell markers are positive, patients are to be sent for upfront chemoradiation. For doubtful cases where markers are equivocal, biopsy or excision is needed.^[3,4,5] All other pineal region tumors which are symptomatic need to be operated with GTE being the desired goal.

Surgical approaches are – supracerebellar infratentorial (SCIT) or Stein’s approach (commonest), occipital transtentorial (OTT) or Poppen’s approach, occipital interhemispheric (OIH), and ETV guided.^[2,5] Stein’s approach is used when the tentorium slope is favorable and Galenic veins are displaced superiorly by the tumor. This approach allows direct trajectory to the tumor and gravity-aided cerebellar retraction and avoids blood pulling in the field. Venous air embolism is the main complication.^[2-5,8] An OTT approach is used when the tentorium is steep, Galenic veins are displaced inferiorly. Transient post-operative hemianopia due to occipital lobe retraction is the main complication with this approach.^[3,4] ETV is mainly indicated where biopsy is needed for a suspected germ cell tumor with concomitant hydrocephalus (for CSF diversion). For patients with coexistent hydrocephalus, pre-operative VP shunt needs to be avoided as much as possible due to shunt-related complications. For CSF diversion in patients presenting with poor GCS score, EVD is preferred with definitive surgery at the earliest once the patient improves.^[4,5] Despite this, a small subset of patients may have residual hydrocephalus after surgery and might need temporary or permanent CSF diversion.

Following microsurgical excision, all specimens are to be sent for histopathology and the decision for adjuvant therapy is to be taken based on the nature and grade of the tumor in the histopathological report.

The histopathological spectra of pineal region tumors are quite variable as well starting from simple pineocytoma and meningioma to pineoblastoma and germ cell tumors.^[2-5,7-9] This affects the surgical outcome as well. For pineocytoma, meningioma and epidermoid cyst; gross or near total excision is possible most of the time whereas for pineoblastoma, grade 3 PPTID safe maximal or subtotal resection is often the achievable and realistic goal for their diffuse and locally invasive nature.^[4,5,10]

In this retrospective observational study, we evaluated 21 patients over 7 years. All except two patients were operated on in the sitting position with GTE/near total excision achieved in 20 patients. Histopathological spectra were studied [Figures 4,5 and 7] and immediate post-operative assessment, as well as 1–5 years follow-up, has been done.

In a published series of 76 pineal region, tumors operated over 18 years^[2] with an age range of 5 months-82 years. About 70% of patients had pre-operative hydrocephalus (85% in our study). Post-operative morbidity was higher in pineal region grade 2–4 gliomas. Disease-reld mortality is 18% in that series which of course is accountable to the higher number of cases and longer follow-up. About 90% of patients were operated on in the sitting/semi-sitting position in the series similar to ours.

Abecassis et al.^[3] published a series of 50 pineal region tumors operated over 10 year time period. They encountered only 2 mortalities in the post-operative period. The SCIT approach was used in 32% of cases whereas the OIH approach was used in 26% of cases. Germinoma was the most prevalent tumor in their series and microsurgical GTE was achieved in only 62% of cases.

Mottolese et al.^[6] published a retrospective study of 31 pineal region tumors operated in a sitting position by infratentorial supracerebellar approach and demonstrated its safety as well as no post-operative mortality and good short- and long-term outcome of both adult and pediatric patients. The median follow-up in their study was 8 years. Pineocytoma and glioma comprised more than half of these cases.

Uda et al.^[11] reported an operated case of the pineal region mixed germ cell tumor in an adult patient. We had two similar patients in our series whose pre-operative germ cell markers were inconclusive but eventually came out to be primitive embryonal tumors [Figure 3].

In a published a series of 23 operated patients of pineal region gliomas,^[12] authors found a favorable survival profile in low-grade glioma cases on average 2 years follow-up. In our study, the only patient with pineoblastoma expired in the post-operative period and two glioma patients did not experience recurrence on the latest follow-up. Magrini et al.^[13] did a retrospective review of pineal region gliomas and showed favorable outcomes in low-grade glioma patients as seen in our study as well. They used the sitting position in 53.5% of the cases. A study by Fedorko et al.^[9] evaluated 20 pineal region tumors operated over 6 years. Pre-operative VP shunt was done in one patient and another four patients after surgery. Short- and long-term follow-up showed favorable QOL profile in operated cases in the study. We found similar results.

In a published a series of 32 pineal region SOLs operated over 20 years,^[8] patients were operated on in the semi-sitting position. The authors found this operative approach safe with no surgical mortality and a favorable post-operative morbidity profile. Pre- and post-operative CSF diversion is similar to our series. Complete excision was achieved in 50% of cases. Yamaki et al.^[14] showed in a review of pineal region papillary tumors that GTE of tumor and smaller tumor size is associated with good prognosis. Nam et al.,^[10] in a review of 17 cases of PPTID, showed no evidence of any definite prognostic factor in these cases and recommended rigorous long-term follow-up.

Shepard et al.,^[4] in their retrospective review of 68 operated patients of pineal region tumors over 25 years with a median follow-up of 95.7 months, showed a 5-year overall survival rate of 70.2%. Most of the patients (81%) in that series were operated by SCIT and OTT approaches like ours. However, the CSF diversion rate was high in their series (77.9%).

CONCLUSION

In this study, we have shown that pineal region tumor surgery in the sitting position is relatively safe with good microsurgical excision results. The pathological spectrum can be versatile, and hence, adjuvant therapy may be needed. In our follow-up, we have favorable outcomes for these patients. However, more rigorous follow-up to determine the long-term outcome is necessary.