Currently, the single nostril transsphenoidal approach has become the preferred option for most pituitary adenoma resections. How to further improve the total tumor resection rate and reduce complications has become a clinical concern. In this paper, we review and summarize the data of 144 surgical cases performed since 2000, and discuss some detailed factors affecting the surgical outcome in combination with the literature.
1.Objects and methods
1.1, General data 67 cases in males and 77 cases in females, aged 17 to 75 years old, with an average age of 37.5 years.
1.2. Clinical manifestations Amenorrhea and lactation were the first symptoms in 65 cases, among which 33 cases were combined with headache, 25 cases were combined with varying degrees of visual acuity loss and visual field defects; 12 cases showed acromegaly, 7 cases showed Cushing’s syndrome combined with hypertension and diabetes mellitus; 22 cases had simple headache; 32 cases had headache combined with visual acuity loss; 5 cases were found by chance; 1 case was due to mental abnormalities, intractable hypotension In one case, pituitary adenoma was detected by cranial imaging due to mental abnormalities and intractable hypokalemia.
1.3. Imaging data All patients were examined by MRI in the saddle area before surgery to determine the tumor morphology, nature and relationship with adjacent structures, and to clarify the anatomical characteristics of nasal cavity, pterygoid sinus and saddle base.
1.3.1. Tumor condition: 37 cases of pituitary microadenoma (diameter <1cm), all tumors were located in the saddle, 1 case combined with part of the empty pterygoid saddle, 2 cases partially wrapped one side of the internal carotid artery; 52 cases of pituitary macroadenoma (diameter 1-3cm), 55 cases of giant pituitary adenoma (diameter >3cm). In the group of pituitary macroadenoma and giant pituitary adenoma, 22 cases were accompanied by tumor stroke and cystic degeneration, and 20 cases had clear cavernous sinus infiltration, of which 7 cases developed to the lateral internal carotid artery.
1.3.2, nasal cavity and pterygoid sinus situation One patient had a combination of bilateral nasal multiple polyps and 17 patients had significant nasal septal deviation. The septum was located in the midline or slightly off the midline in 71 cases; in 33 cases, the septum was located next to the midline; in 14 cases, the septum was double, either on both sides of the midline or one of them was in the midline; in 4 cases, oblique septal shadow was seen, and in 2 of them, the sagittal MRI showed the septum in the midline. In 12 cases, the septum was not seen in the pterygoid sinus because the tumor filled the pterygoid sinus.
In 125 patients, coronal CT of the saddle area was performed within one week after surgery to observe the extent of saddle base resection and to compare with the preoperative images. All patients underwent MR examination of the saddle area within 2 months after surgery to confirm the extent of tumor imaging resection and optic nerve decompression.
1.4. Intraoperative operation All patients underwent single-nostril transsphenoidal sinus approach and pituitary adenoma resection. In one case, a nasal polyp was removed first, and then a transsphenoidal approach was performed two weeks later.
All patients were placed under general anesthesia with tracheal intubation in a supine position, with attention to keeping the face in a horizontal position and, if necessary, a thin pillow under the head to ensure a proper head position. The right nostril approach was chosen, and after filling with ephedrine swabs and dilating the right nasal wing with a small nasoscope, a nasoscope of appropriate size and length was inserted in the direction of the expected anterior wall of the pterygoid sinus to the closest to the butterfly mouth according to the preoperative measurement of the distance and angle between the nostril and the anterior wall of the pterygoid sinus. The extent of saddle base resection is determined according to the relative position of the tumor to the saddle base and the saddle base bone is removed. In some cases, the extent of saddle base resection exceeds the line of the opening of the pterygoid sinus and reaches the level of the anterior bed prominence, with the aim of removing as much of the suprasellar tumor as possible. The dura of the saddle base is incised in the same size and the tumor is gradually scraped away in the direction indicated by the imaging. For the tumors that protrude into the suprasellar area, a step-by-step approach was taken to remove the tumors from the base of the saddle before the suprasellar area, and from the periphery before the center. If the tumor does not descend into the saddle, the anesthesiologist is instructed to perform thoracic pressure breathing and scrape the tumor into the saddle with a spoon.
For those with intra-tumoral hemorrhage, further resection should be attempted if there is residual tumor. For patients with cavernous sinus bleeding or residual tumor section bleeding, we can stop the bleeding by compressing with a thin sheet of gelatin sponge and then fix it with a few drops of otocerebrosides to prevent further bleeding after surgery. For partial resection of tumor with aggressive bleeding in the tumor cavity, a cystic catheter inflatable tube is placed in the saddle to stop the bleeding by compression.
After resection of the tumor, for those without obvious cerebrospinal fluid leakage, a suitable gelatin sponge was cut and filled into the bone window of the saddle base, and after draining the blood, EC gel was injected into the surface of the gelatin sponge quickly. If there is no cerebrospinal fluid leakage, the tumor is partially resected, then only the pterygoid sinus is filled with gelatin sponge. After removing the rhinoscope and repositioning the nasal septum and nasal mucosal flap on the operated side, the nasal cavity was filled with Vaseline oil gauze equally on both sides. In the early stage, multiple short strips of Vaseline oil gauze were used, and in the later stage, all long strips of Vaseline oil gauze were used to end the operation.
2.Results
The degree of tumor imaging resection was determined according to the MRI of the saddle area within 1-2 months after surgery, and the treatment effect was determined by combining with the endocrinological examination and the improvement of patients’ symptoms and signs.
All 37 cases of pituitary microadenoma were completely resected; 52 cases of pituitary macroadenoma were completely resected in 44 cases, and 8 cases were largely resected; 55 cases of giant pituitary adenoma were completely resected in 24 cases, largely resected (more than 90% resection) in 11 cases, and partially resected (50%-80% resection) in 20 cases.
In one case, the intraoperative bleeding was extremely serious, with about 1200 ml of intraoperative bleeding and 700 ml of blood transfusion, and the tumor was so hard and tough that the attempt of total resection was finally abandoned. Postoperative MRI confirmed that the tumor was only partially resected (50%).
Intraoperative saddle septal rupture and cerebrospinal fluid leakage occurred in 12 cases, all of which were sealed by the aforementioned method, and no cerebrospinal fluid leakage occurred after surgery.
In 2 cases, nasal bleeding occurred on the operated side when the oil gauze was removed from the nasal cavity after surgery, and the bleeding stopped 24 hours after re-caulking.
Local postoperative complications associated with the surgery included: fracture of the medial orbital wall-optic nerve canal resulting in optic nerve injury, permanent and severe loss of vision in one eye, loss of smell in 5 cases, and cysts in the pterygoid sinus in 3 cases.
There were 30 cases of postoperative transient enuresis, all of which resolved within one week after surgery with treatment.
3.Discussion
3.1. The importance of detailed reading of preoperative imaging data
At present, for patients with pituitary adenoma, MRI or CT thin coronal scan of the saddle area has become the most basic and mandatory imaging examination [1]. The significance of these examinations, in addition to determining the diagnosis, selecting treatment options, and assessing surgical risk and prognosis, is in the following aspects that provide clear information for transsphenoidal surgery and require special preoperative attention.
The axial and coronal images of the nasal cavity can determine whether the nasal septum is deviated; whether the turbinates are overly enlarged; and whether there are polyps and neoplasms in the nasal cavity. The above information can be used to make a preliminary judgment on which side of the nostril to choose during surgery, the extent of transsphenoidal sinus exposure, and whether abnormal structures in the nasal cavity such as turbinates and nasal polyps need to be treated first. In one of our patients, multiple nasal polyps were found on preoperative imaging, and the patient was first treated surgically by a specialist before undergoing transsphenoidal pituitary tumor resection. The surgical safety and efficacy were ensured.
Measuring the distance and angle of the nostril from the anterior wall of the pterygoid sinus is important to quickly and accurately approach and expose the anterior wall of the pterygoid sinus with a nasal scope of appropriate length and at an appropriate angle. Blindly inserting the rhinoscope into the nasal cavity and propping it up will damage the surrounding normal structures and increase collateral damage. In our case, a fracture of the medial orbital wall-optic nerve canal causing optic nerve injury, permanent and severe loss of vision in one eye and loss of sense of smell in 5 cases were considered to be serious complications caused by fracture of the upper sieve plate of the nasal cavity and extensive damage to the olfactory mucosa due to overly large and deep insertion of the rhinoscope into the nasal cavity. This should be taken as a warning [2] [3].
Careful study of coronal and sagittal scans of the pterygoid sinus should be used to determine the degree of pterygoid sinus development and the position of the tumor in relation to the saddle base, in addition to paying special attention to whether the pterygoid sinus is hypoplastic. Statistics from the literature [4] [5] show that in approximately 10% of patients, the internal carotid artery and optic nerve are located directly under the mucosa of the pterygoid sinus bilaterally, with no bony interval between them. This is particularly likely in overdeveloped pterygoid sinuses. In this case, excessive resection of the bony pterygoid sinus and tearing of the mucosa of the pterygoid sinus increases the likelihood of injury to the internal carotid artery and optic nerve. In such cases, the extent of resection of the anterior wall of the pterygoid sinus and the mucosa of the pterygoid sinus should be limited so that the saddle base is fully exposed. In our surgical practice, in six cases of overdevelopment of the pterygoid sinus, limited resection of the anterior wall and mucosa of the pterygoid sinus was performed to expose the saddle base. Of course, direct visualization using endoscopic guidance would be safer [6] .
Most of the bony septa in the pterygoid sinus are present as a single mediastinum, but some of them are not in the midline. Bony septa within the pterygoid sinus that are complex and variable on MRI or CT require special preoperative and intraoperative attention [7]. In our surgical practice, oblique septa off to one side, multiple septa, or even bony septa within the pterygoid sinus with near horizontal alignment are seen from time to time. Careful determination of the location and alignment of these bony septa and their relative position to the midline, saddle base and tumor based on preoperative images of the pterygoid sinus is important for more accurate exposure and resection of the adequate saddle base and, consequently, the tumor, and should be given adequate attention.
The interpretation of saddle base imaging is of great significance for accurate and complete resection of the tumor while avoiding damage to other important structures. In our case, a patient with pituitary microadenoma combined with a part of empty saddle, part of the saddle base was precisely removed on the basis of careful reading of MRI, so that the tumor was completely removed and intraoperative cerebrospinal fluid leakage was avoided. For huge tumors protruding to the suprasellar, anterior or parasternal saddles, adequate intraoperative resection was performed on the basis of preoperative precise measurement of the width and height of the saddle base bone on MRI or CT films, which was conducive to maximum exposure and resection of the tumor [7] [8] [9]. In this group, for some of the substantial tumors that protrude more to the saddle, the intraoperative saddle base resection range exceeds the line of the opening of the pterygoid sinus and reaches the level of the anterior bed protrusion, aiming to remove as many supra-saddle tumors as possible, which received better results.
3.2. Some details of intraoperative operation
3.2.1. The extent of resection of the mucosa of the pterygoid sinus There is no unanimous conclusion on whether complete or limited resection of the mucosa of the pterygoid sinus, different treatment methods have any effect on the function of the pterygoid sinus and whether it will cause postoperative pterygoid sinusitis or pterygoid sinus cyst [10]. From our three cases of postoperative pterygoid sinus cysts, two cases had incomplete resection of the pterygoid sinus mucosa and one case had complete resection, while we have performed only partial resection of the pterygoid sinus mucosa in about 60 cases during transsphenoidal surgery. Therefore, our limited experience cannot yet provide a causal relationship between the degree of resection of the pterygoid sinus mucosa and the incidence of pterygoid sinus cysts. However, in the case of overdeveloped giant pterygoid sinuses, we emphasize that only limited resection of the mucosa of the pterygoid sinus should be performed to ensure exposure of the saddle base, for the reasons described above.
3.2.2, extent of resection of saddle base bone For giant pituitary adenoma, in order to increase the extent of tumor resection as much as possible, the extent of resection of saddle base bone should be as large as possible without damaging both internal carotid arteries and optic nerve and optic cross. The extent of resection should be determined microscopically with the actual spacing of the forceps tips in addition to the imaging cues to avoid misjudgment of the area due to the magnification effect of high magnification microscopy. In the case of pituitary macroadenoma, a moderate window can be opened above the line of the pterygoid sinus opening. The use of endoscopy and navigation systems facilitates safe and adequate saddle base resection [8] [9].
3.2.3. Resection of giant tumors For giant tumors, they should be resected or scraped sequentially under the microscope from the saddle base to the saddle, in all directions. In particular, the sequence of resection of the saddle base part, then the middle part and finally the upper part of the saddle should be emphasized, which is one of the effective operation methods to ensure the maximum resection of the tumor [10] [11]. However, the transsphenoidal total resection rate of some giant pituitary adenomas is still low due to their infiltration in multiple directions or their stiff and fibrotic texture. The currently adopted methods of increasing intracranial pressure, enlarging the transsphenoidal approach, and endoscopic direct vision or navigation guidance for suprasellar and parasternal tumor resection [12] [13] [14] have all improved the resection rate of such tumors to varying degrees.
3.2.4. Prevention and management of intraoperative bleeding Intraoperative bleeding in transsphenoidal pituitary adenoma resection can originate from the nasal cavity, the bones and mucosa of the pterygoid sinus, the dura and cavernous sinus of the saddle base, the tumor and the internal carotid artery [15].
For bleeding from the bone and mucosa, first of all, gentle manipulation under direct vision during exposure and resection should be emphasized, and rough twisting and tearing are prohibited; mucosal bleeding can be electrocoagulated or compressed, and bone bleeding can be closed with bone wax.
Intraoperative incision of the saddle base dura should be preceded by observation of the presence of dark blue venous sinus cavities in the saddle base dura within the proposed incision area under high magnification microscopy [15]. Simple dural hemorrhage can be treated with weak current electrocoagulation, while hemorrhage from the cavernous sinus should be compressed with gelatin sponge and reinforced with EC ear brain glue if necessary.
Bleeding during tumor resection has firstly the factor of abundant tumor blood flow, followed by bleeding from the residual tumor section, and also venous bleeding after resection of the tumor invading the cavernous sinus and the most serious bleeding from internal carotid artery injury. Total tumor resection is the most effective means to reduce tumor bleeding, and accurate electrocoagulation and compression can also be used to stop bleeding. The filling of the saddle with various hemostatic devices must be moderate. Excessive occlusion may cause severe visual impairment, while saddle base occlusion without effective hemostasis of intra-saddle hemorrhage may cause acute intra-saddle hematoma, which also causes severe visual impairment [15] [16] [17] [18] [19]. In our group of cases, patients with more intra-saddle hemorrhage were sparingly filled with thin sheets of gelatin sponge or Surgecel for intra-saddle and pterygoid sinus and fixed with EC ear cerebral glue. Unless there was a clear cerebrospinal fluid leak, the saddle base was generally not closed to prevent blood accumulation in the saddle. one case of partial tumor resection with aggressive hemorrhage in the tumor cavity was placed in the saddle with a bladder catheter inflatable tube, which also had a drainage effect while compressing the hemorrhage [15].
Hemorrhage from the internal carotid artery is highly likely to be catastrophic. It can be damaged during resection of the anterior wall of the pterygoid sinus and the bones and mucosa of the saddle base, and during resection of tumors. Therefore, exposure and gentle manipulation within the safe range strictly according to imaging cues, and forcible pulling and rotation with the spatula in the presence of resistance during scraping of the tumor in the pars anterior to the saddle and cavernous sinus are strictly forbidden to avoid injury to the internal carotid artery. In case of severe bleeding, only tight compression and tamponade should be performed [15].
3.2.5. Intraoperative management of cerebrospinal fluid leak due to saddle septal rupture during transsphenoidal pituitary tumor resection is difficult to avoid completely. Sometimes it is even a dilemma for the surgeon to choose between total removal of the tumor or avoiding cerebrospinal fluid leakage. There are many reports on methods to seal and reconstruct the saddle base after resection of tumor to stop cerebrospinal fluid leakage [20] [21]. In our group, all 12 cases of intraoperative cerebrospinal fluid leakage were tightly filled with gelatin sponge or fat block plus EC ear brain gel to the dura and pterygoid sinus of the saddle base, plus the postoperative head elevation of 30-45 degrees and moderate extension of nasal filling time to 4-5 days after surgery, which effectively avoided the occurrence of postoperative cerebrospinal fluid nasal leakage.
3.2.6. Correct filling of Vaseline oil gauze As the last step of surgery, correct filling of Vaseline gauze should not be neglected. It should be filled under the principle of bilateral equalization under microscopic observation after the nasal septum and nasal mucosal flap are repositioned to ensure accurate compression of mucosal bleeding points and promote mucosal repositioning and healing, and to avoid septal deviation. In two cases of this group, nasal bleeding occurred on the operated side when the oil gauze was withdrawn from the nasal cavity after surgery, and it stopped after 24 hours of re-caulking. The reason was that the mucosal flap was still separated from the septal bone and the oil gauze was withdrawn with too much movement. In addition, in the early stage of filling the nasal cavity with a short strip of oil gauze, there was a situation that the oil gauze was left in the nasal cavity and was difficult to be removed.
It has been proved that most pituitary adenomas can be treated satisfactorily by butterfly surgery. Preoperative detailed reading of the imaging data and careful operation of each step during surgery are essential to improve surgical efficacy and prevent and reduce complications.