Pituitary adenoma is a common disease in neurosurgery. Because of its concealed location and many adjacent important structures, the tumor is not easily excised completely and has a high chance of recurrence. Currently, the transnasal butterfly approach is the first choice for the surgical treatment of most pituitary adenomas. With the popularity of transnasal butterfly surgery, the number of patients who need to undergo reoperation for recurrence after surgery is gradually increasing.
Whether the previous surgery was performed microscopically or endoscopically, the surgical pathway is significantly affected, making reoperation significantly more difficult and risky. With the rapid development of skull base endoscopic technology in recent years, endoscopic transnasal butterfly reoperation for pituitary adenoma has become a more common surgical approach. A preliminary analysis of the endoscopic transnasal butterfly approach for reoperative resection of recurrent pituitary adenoma is presented to summarize its characteristics.
1.Data and methods
(1) Clinical data
From January 1, 2008 to June 30, 2012, a total of 527 pituitary tumor surgeries were performed, including 52 cases of endoscopic reoperation. All reoperated patients had undergone transnasal pterygoid pituitary adenoma resection for pituitary adenoma, of which 18 cases (34.62%) were endoscopic and the rest were microscopic. All were pathologically confirmed as pituitary adenomas after the initial surgery. All patients underwent reoperation for recurrent pituitary tumors diagnosed preoperatively.
There were 25 males and 21 females with a mean age of 44.08 ± 13.69 years. Among them, 42 were operated twice, 42 cases. Those who had three surgeries or more, 4, 10 cases. The time of reoperation ranged from three months to 11 years since the last surgery, with a mean of 3.96±4.8 years. There were 32 cases with visual field defects before surgery; 29 cases had endocrine abnormalities, including 12 cases with PRL more than twice the normal value, 3 cases with elevated growth hormone, 6 cases with decreased progesterone, 6 cases with decreased thyroid stimulating hormone, and 7 cases with decreased adrenocorticotropic hormone. All of the prolactin adenomas were those who were ineffective in taking bromocriptine or could not tolerate the drug.
2. Methods
All cases were performed under general anesthesia with a single nostril endoscopic transnasal pterygoid approach for tumor resection. The patient was placed in the supine position under general anesthesia, and the patient was fixed with a Mayfield head frame in cases requiring intraoperative navigation. 0.5% dilute iodophor was used to disinfect the nasal cavity. The nasal cavity was explored through the right single nostril with a 30° endoscope and the adhesions of the nasal mucosa were released. According to the nasal septum and middle turbinate between find the pterygoid sieve crypt, the pterygoid sinus opening sometimes adhesions can not be seen, after confirming the location can be cut mucosa to expose the location of the pterygoid sinus opening.
The mucosa and scar tissue in the pterygoid sinus are removed strictly in accordance with the midline, and the dura of the saddle base is opened between the bulge of the internal carotid artery on both sides, and the tumor is removed under direct endoscopic view. After resection, the tumor cavity was filled with Surgicel hemostatic fiber and gelatin sponge, and the skull base was reconstructed with double-layer artificial dura and otocerebral glue. Cranial CT examinations were performed from 8 to 24 hours after surgery. If cerebrospinal fluid nasal leakage was found intraoperatively, muscle and fat fillings of the lower limbs were taken and lumbar pool drainage was performed at the same time. Those with postoperative cerebrospinal fluid leakage were given lumbar pool drainage and bed rest.
3.Results
(1) There were 33 cases (63.46%) of total tumor resection, 11 cases (21.15%) of subtotal resection and 8 cases (15.38%) of partial resection. Intraoperative nasal leakage of cerebrospinal fluid occurred in 8 cases (15.38%) and within 48 hours after surgery in 3 cases (5.77%). All of them were cured within one week by lumbar pool drainage and bed rest.
(2) Postoperative pathology, according to the immunohistochemical results: 33 cases of non-functional adenoma, 13 cases of prolactin adenoma, 4 cases of growth hormone adenoma, 2 cases of ATCH adenoma, and 2 cases of pathology without tumor.
(3) Postoperative visual field improved significantly in 8 cases, no improvement in 4 cases, and no deterioration cases. Preoperative PRL was elevated in 12 cases, and postoperative PRL was still elevated in 5 cases, two of which were significantly lower than before surgery. Growth hormone was elevated in 3 cases, normalized in 1 case after surgery, and decreased significantly in 2 cases. Progesterone decreased in 6 cases, thyroid stimulating hormone decreased in 6 cases, and adrenocorticotropic hormone decreased in 7 cases, all of which did not improve significantly after surgery.
4. Discussion.
(1) Preoperative diagnosis.
In this group, there were 2 cases in which no tumor cells were seen in the pathology report after surgery. It was considered that the postoperative scar growth was misdiagnosed as pituitary tumor recurrence. After pituitary tumor surgery via nasal butterfly, scar hyperplasia appears in the pterygoid sinus and saddle, which may sometimes be mistaken for tumor recurrence, and it is not easy to distinguish it by preoperative imaging. It is recommended to pay attention to timely imaging and endocrinology review after pituitary tumor surgery to keep complete data and determine whether the lesion is recurring according to the progress of the lesion. For recurrent cases with slow changes and no obvious clinical symptoms, surgery is advisable with caution. Although the timing of MRI review after pituitary tumor surgery is still controversial, most authors agree that about three months after surgery is generally appropriate. Within three months, changes in the ischemic and edematous pituitary tissue and saddle environment alter the signal characteristics of the pituitary and tumor.
The residual tumor, inflammatory changes, and normal pituitary gland are not easily distinguished due to incomplete absorption of the saddle filler and local tissue inflammation. However, there is no unanimous opinion on the interval of MRI. Meanwhile, for functional pituitary adenoma, the endocrine changes need to be detected, and those with obvious recurrence of endocrine indexes should be reviewed as early as possible for timely radiotherapy. In one case of GH adenoma in our group, GH decreased after the first surgery, and then increased within three months, and there was no significant improvement after radiotherapy, and the tumor was found to recur after three months on reexamination of MRI. The tumor recurred again six months after reoperation.
(2) Structural changes of nasal cavity.
The mucosa of nasal cavity was adherent and normal anatomical landmarks were unclear. Most patients had obvious mucosal adhesions in the nasal cavity, and some patients had perforated nasal septum. Especially, the perforation at the root of the nasal septum is easily confused with the opening of the pterygoid sinus, leading to misdirection. Domestic surgeons rarely use the operation of removing the middle turbinate, so most patients can be guided by the middle turbinate to find the opening of the pterygoid sinus. In some patients, the mucosa of the nasal cavity is significantly atrophied after the previous operation, and the endoscopic field of view is enlarged compared with the initial operation, so it is easy to establish the working channel of the nasal cavity.
(3) Bony structural changes
The bony structures have been destroyed after the previous surgery, and there are various changes when reoperation is performed, mainly.
Bony nasal septum is incomplete. The degree of nasal septal defect varies according to the different operating habits of the previous surgeon. In this group of reoperations, none of the bony nasal septal roots were intact. Intraoperatively, the nasal septal root is an important midline positioning marker. In patients with partial absence of bony nasal septum, the endoscope can easily expand the operative field slightly anteriorly and inferiorly, which can mostly effectively reveal the edge of the previous surgery and find the residual nasal septal root, which is difficult to do with microscopic surgery. This method is particularly effective in those whose previous surgery was microscopic, where the septal defect is limited in extent because of the restricted operative field.
The wall and septum of the pterygoid sinus and the condition of the saddle base are highly variable. In this group, in two cases, the bony structures of the anterior pterygoid sinus wall and the pterygoid sinus septum were seen to be almost completely restored intraoperatively, but because they were new bone, thin and tough, and adhered tightly to the mucosa, the feel was obviously different when grinding and drilling to remove them, and attention should be paid to them during surgery. Some patients have obvious hyperplasia in the wall of the pterygoid sinus, but there are still obvious defects. At the same time, because of the presence of a large amount of scar and adherent tissue in the nasal mucosa and pterygoid sinus, the anatomical relationship is confused, which can easily cause difficulties in judgment. At this time, it is necessary to patiently clear the scar tissue and then circumferentially expand the scope of the bone window according to the situation until the surgical field is satisfactorily revealed. During the treatment of the pterygoid sinus, it is crucial to perform a thin layer CT scan of the pterygoid sinus before surgery to carefully interpret the changes in the bony structures.
The advantage of endoscopy compared with microscopy in the operation stage in the pterygoid sinus is that the field of view is wide, the viewpoint can be converted at will, and most of them can clearly expose the structures such as the optic nerve bulge on both sides, the internal carotid artery bulge, and the internal carotid artery trap of the optic nerve, and the operation of the open saddle base is performed under direct vision. This facilitates localization and avoids damage to important structures. In cases with particularly complex structures, the use of intraoperative navigation, if available, can be very effective in helping with intraoperative localization and avoiding disorientation and misinjury. If intraoperative disorientation occurs, the operation should be stopped and imaging should be performed to clarify the localization before continuing the operation, and blind exploration is contraindicated. When approaching the saddle base, if the structure is unclear, micro Doppler ultrasound can be used to probe the position of the internal carotid artery when available.
(4) Saddle base operation: After the previous operation, some patients have new bone, the saddle base bone hole becomes smaller, and some of them completely regain the closed state. It needs to be removed again with high speed grinding drill. Before this operation, we must remove the scar tissue in the pterygoid sinus, clearly reveal the structure in the pterygoid sinus, confirm the border of the bone window after the last operation, and use the internal carotid artery bulge, optic nerve bulge and slope cut as reference to determine the scope of saddle base bone grinding.
In most of the patients, there is no bone in the saddle base, so if necessary, only the previous surgery should be enlarged appropriately. The dura mater of the saddle base should be dissected and the artificial dura mater placed in the previous surgery should be carefully removed in some patients. The endoscope is better than the microscope in this step to distinguish the dura mater from the scar tissue at the base of the saddle.
(5) Tumor resection: In reoperative patients, the structure of the saddle is complicated, and many of them are mixed with scar tissue, and the tumor cavity is irregular in shape. Endoscopy can play the innate advantage of close observation here, which is obviously helpful to reduce the dead angle of observation and operation and increase the probability of tumor resection. In our group, there were 12 patients with residual tumor after microscopic surgery, and 9 patients (75%) were completely resected by reoperation. In contrast, there were 5 patients with residual tumor after endoscopic surgery, and only 2 patients (40%) were completely resected by reoperation. This indicates that the resection effect of endoscopic surgery is better than microsurgery in the first operation. The endoscopic surgery is clearly observed for the tumors invading the cavernous sinus, and many tumors in the dead space of the microscope can be easily removed using endoscopy.
Even in the case of cavernous sinus bleeding, it can be handled with ease under direct vision. The scope of application of transnasal butterfly surgery has been greatly expanded. When dealing with tumors closely related to the cavernous sinus and internal carotid artery, it is important to pay attention to the adjacent relationship. Rely on preoperative MRI data to guide the direction and scope of exploration. Navigation guidance can be performed if available. Ultrasound Doppler is extremely important to determine the location and distance of the internal carotid artery to assist and avoid injury. In some cases in our group, the tumor invading into the cavernous sinus was completely removed intraoperatively, and the internal carotid artery was clearly exposed intraoperatively. This is the result of the combination of the advantages of endoscopic technology and the operator’s excellent surgical skills and rich surgical experience.
(6) Skull base repair.
Re-operated patients, because of the scar growth of the tissue in the operated area, poor flexibility and poor blood supply. Intraoperative and postoperative cerebrospinal fluid leaks are likely to occur, and the healing ability is poor after the occurrence of cerebrospinal fluid leaks. Therefore, more attention should be paid to skull base reconstruction in reoperative patients. Once a suspected cerebrospinal fluid leak occurs intraoperatively, muscle fat should be taken for filling and lumbar pool drainage should be performed. In this group, artificial dura mater and otocerebral glue were mainly used as repair materials.
In some patients, intraoperative cerebrospinal fluid leakage is not seen, but it can appear within 1-2 days after surgery. This may be because the cerebrospinal fluid leak was already present intraoperatively but was overlooked because of the small amount. It is also possible that in some patients, the saddle septal foramen is large after tumor removal and brain pulsation tears the arachnoid membrane causing postoperative cerebrospinal fluid leakage. Therefore, postoperative observation of cerebrospinal fluid leakage should be equally important, and once it appears, prompt lumbar pool drainage should be performed.
Compared with microscopic surgery, endoscopic transsphenoidal reoperation for recurrent pituitary adenoma has many unique advantages: large field of view, flexible change in perspective, clear reveal, favorable intraoperative localization, reduced disorientation, and more safety. The endoscope can penetrate deep inside the narrow channel, with a good view of the deep operation, and can easily increase the degree of tumor removal with less trauma.
However, the disadvantages of endoscopic surgery are equally obvious: firstly, the lack of stereoscopic vision, the image fisheye effect is obvious, and the operation depth completely relies on depth of field as a reference. Secondly, endoscopic surgery cannot be observed behind the objective lens, and there is a blind area on the way to the instrument, which increases the operation risk. The axis of endoscopic operation with observation angle is not consistent with the axis of instruments, which is difficult for beginners to adapt. Due to the special observation and operation characteristics of endoscopy, a lot of specialized training and clinical accumulation are needed to complete endoscopic surgery skillfully, and the most important barrier to promotion is still the training problem.