In this study, the usefulness of the transsphenoidal endoscopic approach in the treatment of a Rathke's cleft cyst is reported.
Between June 1998 and December 2002, 22 patients affected by sellar-suprasellar Rathke's cleft cysts were treated using a transsphenoidal endoscopic approach. Fourteen patients experienced pituitary dysfunction (64%), five experienced visual impairment (23%), and three reported headaches (14%).
The patient follow-up ranged from 8 to 60 months (mean, 33 mo). Pituitary function was restored only in the four patients with hyperprolactinemia, whereas visual impairment and headache improved in all patients. However, when present before surgery, hypopituitarism was unaffected by surgery. Two patients experienced permanent diabetes insipidus (one of them before surgery). Only one recurrence was observed in a 13-year-old girl at 12 months after surgery, and it was treated using a new transsphenoidal endoscopic approach.
The transsphenoidal endoscopic approach represents a straightforward and mini-invasive approach for the drainage and biopsy of a Rathke's cleft cyst.
Rathke's cleft cysts are benign congenital lesions believed to derive from remnants of Rathke's pouch. Nonetheless, some authors have suggested that the origin may be endodermal cells, neuroepithelial cells, or metaplastic anterior pituitary cells (7,23). These cysts can be encountered in 12 to 33% of normal pituitary glands in routine autopsies (3,20,24). Moreover, the introduction of magnetic resonance imaging undoubtedly has improved the possibility of diagnosing such cysts before surgery or incidentally (26). These cysts generally are sellar, although in one-third of cases, they can have a suprasellar component. However, the presence of an entirely suprasellar cyst is exceedingly rare (2). Histologically, a Rathke's cleft cyst is composed of a fibrous wall lined with a single layer of ciliated cuboidal or columnar epithelium.
The treatment of Rathke's cleft cysts mainly is surgical, and the use of radiotherapy for their management remains anecdotal (2,15). A transsphenoidal approach, together with the drainage of the cyst accompanied by a biopsy of the cyst wall, usually is the preferred treatment. However, from the literature, it seems that the use of a craniotomy generally is reported for the treatment of those inaccessible lesions located entirely in the suprasellar region (2,19,25).
In this study, the experience with 22 patients with sellar-suprasellar Rathke's cleft cysts treated between June 1998 and December 2002 using a transsphenoidal endoscopic approach is reported. To our knowledge, this represents the largest series reported in the English language literature of Rathke's cleft cysts treated using a transsphenoidal endoscopic approach.
PATIENTS AND METHODS
Between June 1998 and December 2002, 22 patients affected by sellar-suprasellar Rathke's cleft cysts were treated using a transsphenoidal endoscopic approach at the Department of Neurosurgery of Bellaria Hospital in Bologna, Italy. There were 8 male patients and 14 female patients. Their ages ranged from 9 to 63 years (mean, 37 yr). Fourteen patients (64%) experienced pituitary dysfunction (Table 1). Nine patients experienced hypopituitarism (associated with diabetes insipidus in one patient), four patients had hyperprolactinemia, and one patient had impaired growth. Five patients had only visual impairment (23%), and three patients had only headaches (14%). None of the patients reported in this series was asymptomatic. One patient, a 13-year-old girl, previously was treated with two microscopic transsphenoidal approaches to treat her cystic lesion. All of the patients were assessed radiologically by magnetic resonance imaging and computed tomographic scans. Fourteen lesions were sellar, whereas eight were sellar-suprasellar. The size of the cysts ranged from 10 to 26 mm (mean, 20 mm).
In 16 patients, the transsphenoidal endoscopic approach consisted of the drainage of the cyst and the biopsy of its wall (Figs. 1 and 2). Plastic repair of the sella was performed only in the presence of a cerebrospinal fluid (CSF) leak. In six patients (27%), it was possible to carry out a radical resection of the cyst wall. A CSF leak was observed during surgery in six patients, and it was subsequently repaired by plugging the sella using abdominal fat. Therefore, it was possible to perform marsupialization of the cyst only in 10 patients.
The follow-up of the patients ranged from 8 to 60 months (mean, 33 mo). In two patients, a postoperative CSF leak was observed that required repair using fat via an endoscopic approach. One of these CSF leaks occurred in a 13-year-old girl and was complicated by meningitis. Visual impairment and headache resolved in all patients (Table 2). Pituitary function was restored only in the four patients with hyperprolactinemia. Furthermore, when present, hypopituitarism was unaffected by surgery. Two patients experienced permanent diabetes insipidus; in one patient it was present before surgery and was associated with hypopituitarism, whereas in the other patient, it was caused by the surgical resection of the pituitary stalk to achieve complete surgical resection of the lesion, which had been erroneously diagnosed as craniopharyngioma at frozen section analysis. Only one patient reported transitory diabetes insipidus occurring in the perioperative period. In general, the mean hospital stay was 3 days.
The histological results showed an epithelial lining formed by a single layer of columnar or cuboidal epithelium in 78% of the patients, whereas a pseudostratified squamous epithelium and a mixed cell epithelium were found in 18% and 4% of patients, respectively. In a 49-year-old patient, the cyst was complicated by an abscess that appeared to be sterile after Gram staining.
Only one recurrence was observed, which occurred in a 13-year-old girl at 12 months after surgery. She was treated using a new transsphenoidal endoscopic approach. The patient remained free of any recurrence for 17 months; however, at the last enhanced magnetic resonance imaging scan, the Rathke's cleft cyst seemed to have recurred again. Presently, this patient is awaiting revision surgery.
Rathke's cleft cysts are benign congenital lesions representing 6.6% of all pituitary neoplasms treated using a transsphenoidal endoscopic approach at our institution (unpublished data). The local mass effect seems to be the most important cause of symptoms produced by Rathke's cleft cysts. In fact, these cysts can produce either headaches resulting from their compression on the sellar dura or pituitary dysfunction and visual impairment resulting from their compression on the pituitary gland/stalk or optic chiasm. From the literature, it seems that the most common pituitary disturbances include hyperprolactinemia, diabetes insipidus, and partial or total hypopituitarism (5,14,19,25). However, uncommon presentations include abscess formation within the cyst, aseptic meningitis, and impaired growth (5,17,24). In our series, 14 (64%) of 22 patients had pituitary dysfunction. For this reason, we agree with Shin et al. (22) regarding the presence of a higher incidence of pituitary dysfunction in patients affected by Rathke's cleft cyst than previously reported. Visual disturbance was the second most common symptom in this series (23%), which was also observed by Voelker et al. (25) in their analysis of 155 patients. In this study, the incidence of headache as a presenting symptom was not as high as in other reports (8,10,19). Abscess within Rathke's cleft cyst has been described by several authors (9,10,16,25), and it also was encountered in one patient in this series. Furthermore, we encountered a case of pituitary apoplexy, a rare complication that was described previously only by El-Mahdy and Powell (5).
Surgical treatment using a transsphenoidal approach with marsupialization and drainage of the cyst in addition to biopsy of its wall seems to be the mainstay of treatment (2,3,5,10,13,16,19,25). In general, the recurrence rate after surgery for the management of Rathke's cysts varies from 0 to 33% (5,16). The craniotomy route should be reserved only for inaccessible suprasellar lesions, although Kim et al. (11) described a transsphenoidal supradiaphragmatic intradural approach for the removal of a suprasellar Rathke's cleft cyst confined to the pituitary stalk. Moreover, aseptic meningitis caused by the leakage of the cyst contents within the subarachnoid space may represent a complication of the craniotomy route (21,24). Landolt and Zachmann (13) outlined the effectiveness of the transsphenoidal approach for the treatment of intrasellar craniopharyngioma or Rathke's cyst because of its low rate of recurrence, the low risk of surgical mortality, and the insignificant incidence of complications when compared with that of craniotomy. Recently, a few reports have outlined the usefulness of the transsphenoidal endoscopic approach for the treatment of Rathke's cleft cyst (1,10).
In general, endoscopy, having good magnification and angled vision, permits thorough visualization, especially in the distal portion of the operative field, and also offers the possibility of either operating bimanually through an endoscopic holder or performing a “freehand exploration” of the surgical cavity and surgical field, generally at the end of the procedure. This exploration can also be useful in assessing the entity of the resection, in excluding any CSF leaks, and in repairing the surgical defect. Furthermore, the endoscopic technique is minimally invasive surgery and has low morbidity; thus, it does not influence the parameters of preoperative nasal ventilation and offers the possibility of repetition in cases of recurrence (4).
We agree with Ross et al. (19) regarding their recommendation of mainly carrying out a simple drainage of the cyst via the transsphenoidal route together with biopsy of the cyst wall, without entering the subarachnoid space or damaging the normal structures. Entering the subarachnoid space produces a CSF leak that must be repaired, usually by packing the sella with fat. Kasperbauer et al. (10) found a higher rate of recurrence of Rathke's cleft cyst in patients who had undergone packing of the sella for repair of a CSF leak or the chiasmal support (chiasmapexy). In our series, a CSF leak was observed and repaired using abdominal fat in eight patients (36%). In six patients, the leak occurred during surgery, whereas in two patients, it occurred after surgery. Only two cases of CSF leak occurred in the group of patients who underwent a radical removal of the cyst wall. Nonetheless, only one patient among those undergoing repair of a CSF leak experienced a recurrent Rathke's cleft cyst. This finding may be explained by the short follow-up of the patients reported in this series, although the biological aggressiveness and tendency of the cyst to recur in this 13-year-old patient also may be considered a plausible cause of recurrence. Radical removal of the cyst was accomplished in six patients (27%) in this series. Nevertheless, radical removal was not attempted whenever a portion of the cyst adherent to the surrounding structures could damage neural structures such as the pituitary gland/stalk, the hypothalamus, or the optic chiasm. We believe that extensive removal of cyst wall should be carried out only where it is possible to do so without causing additional and unnecessary pituitary damage. Moreover, the extensive removal of the cyst wall should be carried out only when loose adhesions allow an atraumatic peeling from the pituitary gland and, above all, from the pituitary stalk. Generally, simple drainage of the cyst does not produce any new endocrine deficit. However, in one patient in this series, analysis of the intraoperative frozen section suggested a craniopharyngioma, so we decided to perform an aggressive removal of the cyst wall that caused pituitary stalk interruption, owing to the cyst's tenacious adhesions to the surrounding tissues. Pituitary dysfunction was recovered only in the four patients with hyperprolactinemia (28.6%). This result agrees with that reported by other authors (5,12,22). When present before surgery, pituitary insufficiency was unaffected by surgery. However, visual impairment and headache improved in all patients, and which also has been pointed out by several authors (5,10,19,22). Keeping in mind the possibility of having a patient who has hypophysitis with hypopituitarism resulting from the inflammation produced by the spillage of the cyst contents within the subadjacent hypophysis (6,18), the surgical management of Rathke's cleft cyst becomes opportune even in the presence of subtle symptoms. For this purpose, the transsphenoidal endoscopic approach represents a straightforward and mini-invasive approach for carrying out the drainage and biopsy of Rathke's cleft cyst.
In this study, the effectiveness of the transsphenoidal endoscopic approach for the treatment of 22 cases of sellar-suprasellar Rathke's cleft cysts is reported. To our knowledge, this represents the largest series reported in the English language literature of Rathke's cleft cysts treated using this mini-invasive approach. This mini-invasive approach confirmed an efficacy similar to that obtained through transsphenoidal microsurgery, also with the advantage of a better tolerance and a lower complication rate (Table 3). In all cases, headache, visual impairment, and hyperprolactinemia improved after surgery. However, pituitary insufficiency was unaffected by the surgical treatment.
Because of the risk of serious complications resulting from the variable clinical behavior of Rathke's cleft cyst, surgical treatment, even in the presence of subtle symptoms, is indicated. Because of its low morbidity, the transsphenoidal endoscopic approach can assist us in making a decision in favor of surgical management of these lesions rather than adopting a wait-and-see approach.
This is an interesting report that adds further to the Rathke's cleft cyst experience. We all agree that the transsphenoidal route is best, whether with the microscope or the endoscope. The difference, I believe, is insignificant in time in the hospital and in morbidity. I agree with the authors that a less aggressive approach with drainage and marsupialization of the cyst into the sinus is best. It has been my experience that with more aggressive resection of the capsule, there is a higher incidence of pituitary dysfunction. It is also interesting that the return of preoperative pituitary deficiency is as low as it is. This point may speak to earlier drainage when headache is the sole complaint.
Kalmon D. Post
New York, New York
This is a beautifully presented series of endoscopic approaches for the complicated and controversial problem of Rathke's cleft cyst. The authors have presented excellent initial results and a lucid description of their technique and of the many issues that still concern those of us who treat Rathke's cleft cysts. It is clear that their series includes a higher percentage of large cysts than many others, and this is reflected by the number of patients who have visual loss as part of their presenting complaints. The results with regard to visual loss are superb. The authors have generally taken a conservative approach with fenestration of the cyst and removal of the contents and thus far have had very good outcomes. They do note that those patients who have hypopituitarism preoperatively do not ordinarily recover, and this has been our experience as well. Part of this is related to the inflammatory response that often occurs in the normal pituitary gland and results in hypopituitarism.
Because the recurrence rate in our series had been rather high, approximating 8% in patients followed up for 5 years or more, we have adopted a more aggressive surgical strategy in patients who are older and not in the reproductive years. We reported our series of 78 patients at the 2004 American Association of Neurological Surgeons meeting in Orlando, FL, all of whom were treated via a microscopic transsphenoidal approach over a comparable follow-up period (mean, 32 mo). Given the sizable lesions treated by the authors, they report a surprisingly low number of patients with complaints of headache compared with other published series as well as our own experience, which included 86% of cases. A second unexpected result was the number of CSF leaks encountered (6 of 16) when mere cyst drainage was the goal. When radical cyst wall resections were performed in our series, we encountered a similar rate of CSF leaks, but when simple drainage and biopsy were performed, only 17% of patients experienced a leak necessitating sellar packing. This brings us to a third disparity. The authors cite Kasperbauer et al. (1), reporting a higher recurrence rate after packing of the sella. The authors experienced only one recurrence in their series, a child who experienced a CSF leak and subsequently received sellar packing, thus a 13% recurrence rate in this group. We encountered a recurrence rate of 3% in our group of patients after packing of the sella, compared with a 10% recurrence rate in those who did not require packing, concluding that the recurrence rate is not increased by such action.
We do believe that radical removal of the cyst wall of a Rathke's cleft cyst results in a lower recurrence rate (6% in our series), approximately half the recurrence rate observed after simple drainage and cyst wall biopsy (13%). We have not had excessive difficulty with complications such as postoperative CSF leak (2%) or new hypopituitarism (3%), except for diabetes insipidus (15 new cases), which was most likely the result of aggressive removal of cyst wall adherent to the posterior stalk. The authors appropriately caution those who perform radical approaches that damage to the pituitary stalk is a real possibility, and one would hope that by use of endoscopic techniques, the ability to preserve the stalk might be enhanced.
The issue of whether Rathke's cleft cysts are part of the continuum that ends in craniopharyngioma is not discussed in any detail, but certainly a few cases that suggest that this may be the case have been reported, and we believe that such a continuum exists. We look for further innovations from this excellent group.
Edward R. Laws, Jr.
Adam S. Kanter