Pituitary DOI 10.1007/s11102-014-0587-4
Extracapsular en bloc resection in pituitary adenoma surgery Eui Hyun Kim • Cheol Ryong Ku • Eun Jig Lee Sun Ho Kim
•
Ó Springer Science+Business Media New York 2014
Abstract Purpose A pituitary pseudocapsule often contains tumor tissue and should be removed for radical resection. It can be used as a surgical plane for more radical resection of the tumor in many cases of pituitary adenomas. We evaluated the advantages and disadvantages of extracapsular en bloc capsulectomy. Methods From 1992 until 2011, 1,089 treated patients were grouped according to the resection technique: en bloc capsulectomy, fragmented capsulectomy, or piecemeal resection. Their surgical and endocrinological outcomes and complications were evaluated. Results Extracapsular tumor resection was performed in 263 patients; en bloc capsulectomy in 94 patients and
Electronic supplementary material The online version of this article (doi:10.1007/s11102-014-0587-4) contains supplementary material, which is available to authorized users. E. H. Kim S. H. Kim (&) Neurosurgery, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea e-mail: [email protected] E. H. Kim C. R. Ku E. J. Lee S. H. Kim Endocrine Research Institute, Yonsei University College of Medicine, Seoul, Republic of Korea E. H. Kim C. R. Ku E. J. Lee S. H. Kim Pituitary Tumor Clinic, Yonsei University College of Medicine, Seoul, Republic of Korea C. R. Ku E. J. Lee Endocrinology, Yonsei University College of Medicine, Seoul, Republic of Korea E. J. Lee Endocrinology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA
fragmented capsulectomy in 169, whereas piecemeal resection was performed in 826. Extracapsular resection was performed more frequently in prolactin- and thyroidstimulating hormone-secreting tumors. Total resection was more frequently achieved in extracapsular resection and its chance was 100 % when tumors were removed in an en bloc fashion. For the functioning pituitary adenomas, endocrinological remission was achieved in all patients whose tumors were removed in an en bloc fashion and there was no recurrence. Postoperative cerebrospinal fluid (CSF) rhinorrhea developed in 4.2 and 2.7 % in the extracapsular resection group and the piecemeal resection groups, respectively. The chance of postoperative aggravation of pituitary function was not statistically different between groups. Conclusions Extracapsular resection is critical for radical tumor resection and endocrinological remission. The removal of a pseudocapsule does not increase the risk of postoperative hypopituitarism nor postoperative CSF rhinorrhea. Keywords Capsulectomy En bloc resection Extracapsular resection Pituitary adenoma Pseudocapsule
Introduction Since a pseudocapsule associated with a pituitary adenoma was initially described in the 1900s [1], there have been many reports about pituitary pseudocapsules. These are thought to develop in response to pressure within the pituitary gland itself and become thin multiple layers of compressed pituitary gland tissue containing a reticulin network. It is found in about 50 % of patients with a
123
Pituitary
pituitary adenoma and tends to be more frequent in larger tumors [2]. Histopathology has revealed that tumor tissue is frequently present within the pseudocapsule, which justifies the aggressive resection especially in endocrine-active pituitary tumors [3]. Thus, in recent years, many authors have adopted so-called extracapsular resection and emphasized the importance of a pituitary pseudocapsule as a surgical plane for more radical resection of the tumor [2– 8]. However, most authors have focused on the surgical technique and histological findings rather than on surgical and endocrinological outcomes. It has been our surgical strategy to perform intensive resection of pseudocapsules as we also found they could be a source of tumor recurrence and failure of remission in endocrine-active tumors [3]. Tumor dissection was performed outside the pseudocapsule whenever possible. When extracapsular resection was not possible, tumors were removed in a piecemeal fashion. We grouped our patients based on the resection techniques and evaluated surgical and endocrinological outcomes and complications. The advantages and disadvantages of extracapsular resection are discussed below with a description of our surgical techniques.
Methods Patients From January 1992 until December 2011, 1,372 cases of pituitary adenoma were operated on with a transsphenoidal approach (TSA) at Yonsei University College of Medicine by a single neurosurgeon (SHK). All adenomas were classified according to a modified Hardy radiological classification scheme [9, 10]. Of these, 283 patients were excluded from the study for the following reasons: Hardy type IV tumors, which had cavernous sinus invasion (n = 210); revision surgery for recurred tumors (n = 67); history of radiotherapy (n = 2); and loss to follow-up (n = 4). Thus, 1,089 patients were included. These included 464 male and 625 female patients with a mean age of 43.4 years. The follow-up duration ranged from 1.0 to 18.4 years (mean 4.8 years). The classification of pituitary adenomas according to hormonal activity was verified from postoperative immunohistochemistry of the tumor tissue. Endocrine-active tumors were observed in 537 patients (49.3 %); growth hormone (GH)-secreting tumors were observed in 312, prolactin (PRL)-secreting tumors were observed in 163, adrenocorticotropic hormone (ACTH)secreting tumors were observed in 39, thyroid-stimulating hormone (TSH)-secreting tumors were observed in 15 and both GH- and PRL-secreting tumors were observed in 8 patients. The presence of nonfunctioning (NF) tumors was
123
confirmed in 552 patients (50.7 %). Medical treatment with a dopamine agonist was tried prior to surgery in 80 patients (49.1 %) with PRL-secreting tumors and they showed no or insufficient response. Among the patients with GHsecreting tumors, 19 patients had ever had preoperative octreotide long-acting release (Sandostatin LAR Depot; Novartis Pharmaceuticals Corp., East Hanover, NJ, USA) treatment. For 38 patients with NF pituitary adenomas, medical treatment was tried under the impression of a PRLsecreting tumor, but this failed prior to the surgical treatment. Patient evaluation The preoperative patient work-up included taking a complete history, physical and neurological examinations, and endocrinological and radiological assessments. All patients underwent preoperative dynamic magnetic resonance imaging (MRI) of the sellar region using a 1.5-T system (Signa; General Electric, Minneapolis, MN, USA) or a 3.0-T system (Achieva; Philips, Best, The Netherlands). Immediate postoperative 3-T diagnostic MRI was performed within 48 h after the surgery and the radicality of tumor removal was determined. Follow-up MRI was performed 1 year after surgery and every other year thereafter. The endocrinological assessment with combined pituitary function test (CPFT) was fully described in our previous report [3]. The changes in pituitary function from before and after TSA were classified into four categories: (1) normal to normal, meaning no pituitary hormone deficiency before and after TSA; (2) improved hypopituitarism, meaning the hypopituitarism improved including complete recovery; (3) persistent hypopituitarism, meaning the degree of pituitary hormone deficiency did not change after TSA; and (4) worsened hypopituitarism, meaning the number and/or degree of pituitary hormone deficiency worsened after TSA. The last of these categories included the new development of postoperative hypopituitarism and the aggravation of preoperatively existing hypopituitarism. CPFT was performed to evaluate anterior pituitary function before surgery, 6–12 months after surgery, and every 2 years thereafter. In acromegalic patients, an oral glucose tolerance test (OGTT, 75 g glucose) was performed preoperatively, 1 week after surgery, every 6 months up to 3 years after surgery, and then annually. Serum insulin-like growth factor-1 (IGF-1) was also measured whenever OGTT was performed. Remission was defined when the nadir serum GH level after an OGTT was \1 lg/L with normalization of serum IGF-1 adjusted for age and sex. Patients with Cushing’s disease were evaluated by measurement of 24 h urinary free cortisol and 8 AM serum cortisol level, serum ACTH level, and dexamethasone suppression test, for preoperative diagnosis and determination of remission. The levels of PRL, T3, T4, and
Pituitary
TSH were evaluated in patients with PRL- and TSHsecreting tumors, respectively.
(version 20.0; IBM, Armonk, NY, USA) and P \ .05 was taken as significant.
Surgical procedure Results The procedure was performed in the same way as in conventional microscopic TSA. Tumors were approached differently according to their location and size. When a tumor was not fully visualized or when a remnant tumor was suspicious on an invisible area under the microscope, an endoscope was introduced into the surgical field. After dural opening, we localized the normal pituitary gland first and checked for the consistency of the tumor and the presence of a pseudocapsule. If a tumor showed hard consistency or had a fairly thick pseudocapsule, extracapsular dissection was performed outside it. In this situation, special caution was taken not to dissect outside a normal pituitary gland with improper orientation. If a pseudocapsule covered the entire surface of a tumor, en bloc resection was attempted (en bloc capsulectomy) (Video 1–4). When pseudocapsule formation was incomplete, the tumor was removed as several fragmented pieces (fragmented capsulectomy). When a tumor showed fragile and soft consistency with a scarcely developed pseudocapsule, the tumor was removed in a piecemeal fashion (piecemeal resection). In these cases, any pseudocapsule found over the normal pituitary gland surface was always removed completely and this suspicious tissue was sent to the pathology laboratory for frozen section histopathology. With any faint suspicion of dural invasion, all exposed dura mater was removed as much as possible. When a pseudocapsule was attached onto the overlying thin layer of the normal gland and arachnoid membrane, these were protected with a small piece of Bemsheet (Kawamoto Bandage Material Co., Ltd, Osaka, Japan) and the pseudocapsule was separated very carefully. Even when a pseudocapsule was very adhesive to the arachnoid membrane, gentle manipulation was always mandatory and a small piece of Bemsheet was very useful during this step. Rough pulling of the pseudocapsule was never attempted. Even though dissection was performed very carefully, sometimes cerebrospinal fluid (CSF) leakage encountered. When this was noted during surgery, it was sealed by various direct closure techniques to close the leaking point. Postoperative lumbar drainage of CSF was performed in many of our early cases; however, it has not been used since 2005. Statistical analysis Chi squared and Fisher’s exact tests were used for determining the statistical significance of any differences. Statistical analyses were performed using IBM SPSS Statistics
During surgery, a pseudocapsule was identified in 531 patients (48.8 %). Among these patients, a pseudocapsule was apparent and thick enough to be used as a surgical dissection plane in only 263 (49.5 %). En bloc capsulectomy was achieved in 94 patients and fragmented capsulectomy in 169. The tumors were removed in a piecemeal fashion in 826 patients (75.8 %): 268 with a fragile, fragmented pseudocapsule and 558 with no visible pseudocapsule. Tumor removal fashion according to the type of hormone produced is summarized in Table 1. When compared with the manner of tumor removal in all patients, extracapsular resection was performed more frequently in PRL-secreting tumors and in TSH-secreting tumors whereas NF pituitary adenomas were more likely to be removed in a piecemeal fashion. These differences were statistically significant. Extent of resection The results for the extent of resection are summarized in Table 2. Tumors were removed totally in 258 patients (98.1 %) among 263 patients in whom extracapsular resection was possible. When tumors were removed in an en bloc fashion, total resection was achieved in all 94 patients. In the group with fragmented capsulectomy, total resection was possible in 164 of 169 patients (97.0 %). However, when tumors were removed in a piecemeal fashion, only 93.6 % of patients (773/826) achieved total resection. These differences were statistically significant (P = .004). Unexpected remnant tumors were found on the postoperative MRI in 6 of 826 patients (0.7 %) with a piecemeal resection whereas no patients with extracapsular Table 1 Tumor removal fashion according to the hormonal type of tumor Hormonal type
Extracapsular resection
Piecemeal resection
P value
GH
80 (25.6 %)
232 (74.4 %)
.466
PRL
56 (34.4 %)
107 (65.6 %)
.001
TSH
\.001
11 (73.3 %)
4 (26.7 %)
ACTH
4 (10.3 %)
35 (89.7 %)
.037
GH ? PRL
3 (37.5 %)
5 (62.5 %)
.409
NF
109 (19.7 %)
443 (80.3 %)
.001
Total
263 (24.2 %)
826 (75.8 %)
\.001
ACTH adrenocorticotropin hormone, GH growth hormone, NF nonfunctioning, PRL prolactin, TSH thyroid stimulating hormone
123
Pituitary Table 2 Radicality of tumor removal according to tumor removal fashion
Extracapsular resection En bloc capsulectomy Total resection
98.1 % (258/263) 100 % (94/94)
a
Unexpected remnant tumors
0 % (0/263)
Recurrence
3.1 % (8/258a)
0 % (0/94)
Number of patients with total resection
0 % (0/94 )
Table 3 Endocrinological remission rate according to the hormonal type of the tumor
Extracapsular resection
a
Total number of patients with functioning tumors whose biochemical remission was determined postoperatively
a
P value
93.6 % (773/826)
.004
0.7 % (6/826)
.345
1.3 % (10/773a)
.055
97.0 % (164/169) 0 % (0/169) a
4.9 % (8/164 )
Piecemeal resection
Total
P value
GH
97.4 % (75/77)
88.5 % (193/218)
90.8 % (268/295)
.020
PRL
77.6 % (38/49)
73.1 % (68/93)
74.6 % (106/142)
.564
TSH
100 % (11/11)
75.0 % (3/4)
93.3 % (14/15)
.267
ACTH GH ? PRL
100 % (4/4) 50 % (1/2)
100 % (35/35) 75.0 % (3/4)
100 % (39/39) 66.7 % (4/6)
– 1.000
Total
90.2 % (129/143)
85.3 % (302/354)
86.7 % (431/497a)
.145
En bloc capsulectomy
Others
Total
P value
GH
100 % (29/29)
89.8 % (239/266)
90.8 % (268/295)
.089
PRL
100 % (12/12)
72.3 % (94/130)
74.6 % (106/142)
.037
TSH
100 % (8/8)
85.7 % (6/7)
93.3 % (14/15)
.467
ACTH
100 % (2/2)
GH ? PRL Total
0% 100 % (51/51)
resection revealed any unexpected remnant tumors on their immediate postoperative MRI. However, the difference was not statistically significant (P = .345). When their tumors were totally removed, the recurrence rate 3.1 % (8/ 258) in the group with extracapsular resection, which was a little higher than that of a piecemeal resection group (1.3 %, 10/773) even though it was not statistically significant (P = .055). However, when the tumors were removed in en bloc resection technique, the recurrence rate was zero. Biochemical remission in endocrine-active tumors Postoperative determination of biochemical remission was feasible in 497 of 537 patients with endocrine-active tumors as 40 patients were lost to follow-up. Overall, 431 patients (86.7 %) showed biochemical remission (Table 3). The remission chance was a little higher in an extracapsular resection group (90.2 %, 129/143) than in a piecemeal resection group (85.3 %, 302/354), however, the difference was not statistically significant. When we grouped the patients into an en bloc capsulectomy group and others, as 51 patients with en bloc resection showed a 100 % biochemical remission rate, the patients with en bloc resection
123
Fragmented capsulectomy
Piecemeal resection
100 % (37/37)
100 % (39/39)
–
66.7 % (4/6)
66.7 % (4/6)
–
85.2 % (380/446)
86.7 % (431/497a)
.001
Table 4 Influence of preoperative medical treatment on the tumor removal fashion Extracapsular resection Total number of patients with preoperative medication (n = 137)
49 (35.5 %)
Piecemeal resection 89 (64.5 %)
P value .001
GH (n = 19)
4 (21.1 %)
15 (78.9 %)
.750
PRL (n = 80)
32 (40.0 %)
48 (60.0 %)
.001
NF (n = 38)
12 (31.6 %)
26 (68.4 %)
.276
94 (8.6 %)
826 (75.9 %)
Total number of patients (n = 1,089)
GH growth hormone, NF non-functioning, PRL prolactin
showed much higher chance of biochemical cure than other patients with statistical significance (P = .001). Preoperative medical treatment was attempted in 80 patients with PRL-secreting pituitary adenomas, 19 patients with GH-secreting pituitary adenomas, and 38 NF pituitary adenomas; insufficient responses were documented in all of them before surgical treatment (Table 4). When these 137 patients were classified based on the tumor removal fashion, the patients with preoperative medical
Pituitary Table 5 Incidence of CSF leak according to tumor removal fashion
Intraoperative CSF leak
Extracapsular resection
Piecemeal resection
P value
59.3 % (156/263)
41.6 % (344/826)
\.001
Extracapsular resection
Postoperative CSF rhinorrhea
En bloc resection
Fragmented capsulectomy
4.2 % (11/263) 1.1 % (1/94)
5.9 % (10/169)
Piecemeal resection
P value
2.7 % (22/826)
.211
CSF cerebrospinal fluid
Table 6 Postoperative hormonal outcome according to tumor removal fashion Postoperative hormonal outcome
Extracapsular resection
Piecemeal resection 137 (19.0 %)
Preserved normal pituitary function (n = 189, 19.7 %)
52 (22.1 %)
Improved hypopituitarism (n = 505, 52.7 %)
119 (50.6 %)
386 (53.4 %)
Persistent hypopituitarism (n = 146, 15.3 %)
28 (12.0 %)
118 (16.3 %)
Aggravation of hypopituitarism (n = 118, 12.3 %)
36 (15.3 %)
82 (11.3 %)
235
723
Total (n = 958) a
a
P value
.127
Total number of patients in whom comparison between preoperative and postoperative CPFT was available
treatment were more likely to undergo extracapsular resection (P = .001) compared with tumor removal fashion in total patients. However, when patients were subclassified into each hormonal type of tumors, this difference was observed only in patients with PRL-secreting pituitary adenomas (P = .001).
(12.3 %). When they were classified based on the fashion of tumor removal, no statistically significant differences were found (P = .127).
Cerebrospinal fluid leak
The presence of pseudocapsules was first described in the early 1900s; however, understanding their formation and highlighting their clinical significance have been recent developments. Oldfield and Vortmeyer [2] described how histological pseudocapsule develops by examining the distribution of tumor size in relation to the presence of a pseudocapsule. They emphasized that the pseudocapsule is extremely useful for complete tumor excision. There have been many reports describing surgical techniques using a pseudocapsule as a surgical dissection plane. In those reports, the dissection was performed along the interface between the outer surface of a pseudocapsule and a normal pituitary gland [3–8]. Their techniques increased the complete tumor removal and endocrinological remission rates and decreased the complication and recurrence rates. Qu et al. [6] defined these techniques as pseudocapsulebased extracapsular resection approach. However, this technique is not always possible because whether the resection is performed outside or inside the pseudocapsule largely depends on the consistency of the tumor and the degree of development of the pseudocapsule. In our early experience, we demonstrated that a microsurgical pseudocapsule was found in 55.7 % of patients
Intraoperative CSF leakage was noted in 156/263 (59.3 %) of patients with extracapsular resection and 344/826 (41.6 %) in the group with piecemeal resection (Table 5). The difference in the risks of intraoperative CSF leakage were statistically significant with P values of \.001. Postoperative CSF rhinorrhea developed in 11/263 (4.2 %) of patients with extracapsular resection, which was a little higher than the 22/826 (2.7 %) among patients with piecemeal resection. However, the difference in the risks of postoperative CSF leakage were statistically not significant (P = .211). The possibility of postoperative CSF rhinorrhea was higher in a fragmented capsulectomy group (5.9 %) than in an en bloc resection group (1.1 %) and in piecemeal resection group (2.7 %). Hypopituitarism Comparisons between preoperative and postoperative CPFT were available in 958 patients (Table 6). Overall, normal pituitary function was preserved in 189 patients (19.7 %), improved in 505 (52.7 %), and aggravated in 118
Discussion
123
Pituitary
Fig. 1 Photomicrographs of tumor specimens. a Tumor specimen obtained en bloc with a distinct pseudocapsule during surgery, showing a relatively thick layer of connective tissue at the interface between the pituitary adenoma and the normal pituitary gland;
hematoxylin and eosin (H & E) staining, original magnification 940. b With higher magnification, clusters of tumor cells can be identified in the pseudocapsule; H & E staining, original magnification 9200
with pituitary adenomas [3]. The degree and characteristics of pseudocapsule varied widely: (1) entirely covering the tumor mass; (2) a thin, fibrous envelope; (3) a yellowish, discolored, normal gland-like, thin membrane; (4) thick, fibrous tissue; and (5) sometimes dense fibrosis or calcifications. We believe the tumor size, hormonal type, and the presence of intratumoral hemorrhage can influence the nature of the pseudocapsule. Thus, different surgical techniques should be adopted according to the presence of a surgically resectable pseudocapsule. When a tumor has a fibrous and thick pseudocapsule or when a tumor itself has very hard and fibrous consistency, dissection should be performed outside the pseudocapsule (Fig. 1). By contrast, if the tumor is very soft and fragile and its pseudocapsule is very thin or faint, piecemeal resection is inevitable. However, even in the latter situation, close inspection is essential to check for the presence of any suspicious pseudocapsule over the remaining pituitary gland or arachnoid membrane and this should be removed if it is found.
recurrence, although it was not statistically significant, the recurrence rate was slightly higher in an extracapsular resection group than in a piecemeal resection group, which was mainly because we experienced more recurrent tumors in a group of fragmented capsulectomy. Considering no patients experienced recurrence in the group with en bloc resection, we presume the chance of a small fragment of residue might be higher in cases of a fragmented capsulectomy because fibrosis, calcification, and incompleteness of pseudocapsule were more frequent in this group of patients. Therefore, careful inspection is essential with the suspicion that tumor cells might still reside beyond the main tumor boundary when a pseudocapsule is incompletely developed.
Extent of resection Pseudocapsule-based extracapsular resection is a technique that maximizes the extent of resection, as the tumor is resected completely within its pseudocapsule. The pseudocapsule itself can be used as a reliable boundary of the inner tumor content. In our series, the rate of total resection was much higher in the groups with en bloc resection and fragmented capsulectomy than in the piecemeal group. Although there was no statistical difference in the incidence of unexpected remnants among the three groups (P = .383), no patients who underwent en bloc resection or fragmented capsulectomy showed remnant tumors on their postoperative MRIs. As for
123
Biochemical remission in endocrine-active tumors An increased extent of resection leads to increased remission rates among patients with endocrine-active pituitary adenomas, as demonstrated for GH-secreting pituitary adenomas by Ku et al. [9] and Kawamata et al. [5], and for ACTH-secreting pituitary adenomas by Jagannathan et al. [4] and Monteith et al. [11]. Although this technique is only possible when the tumor has quite a well-developed pseudocapsule, the authors emphasized that the capsule itself contains tumor cells and can be a main cause of persistent hypersecretion of the hormone and possibly the source of recurrence. Also, in our previous reports, we suggested that complete removal of the pseudocapsule is critical to achieve endocrinological remission in patients with endocrine-active pituitary adenomas [3, 9]. However, even in a tumor with a well-developed pseudocapsule, tumor cells might invade the nearby normal pituitary gland and dural layer beyond the pseudocapsule [12]. This is known to be one of the main reasons for discordance
Pituitary
between complete tumor removal and endocrinological remission. For this reason, Laws [12] emphasized that it is necessary to remove a rim of normal pituitary gland surrounding the tumor to improve surgical outcome in patients with endocrine-active pituitary adenomas. However, this fact does not diminish the usefulness of pseudocapsulebased extracapsular resection. This technique enables reliable complete resection of inner tumor contents. Moreover, the pseudocapsule itself might be infiltrated by tumor cells, which justifies the radical removal of the pseudocapsule. Our previous study showed that the presence of a pseudocapsule was more frequent in PRL-secreting pituitary adenomas than in GH- and ACTH-secreting pituitary adenomas [3]. Also, our present study demonstrated that extracapsular resection was more feasible only in PRLsecreting pituitary adenomas. We have already suggested that this might be associated with previous medical treatment with dopamine agonists, which facilitate fibrous tissue formation. Although preoperative medical treatment seems to enhance the formation of a pseudocapsule, it also makes the interface between pseudocapsule and a normal pituitary gland very adhesive, which in turn makes the surgical maneuver quite difficult. As discussed above, we verified that removal of the pseudocapsule increases the extent of resection and endocrinological remission rate. However, because this technique requires more aggressive resection of a pseudocapsule from the surface of an underlying pituitary gland and arachnoid membrane than the piecemeal resection technique, it could increase the possibility of complications. We also tried to clarify our concerns about complications, which are as important as the radicality of tumor removal. Cerebrospinal fluid leakage CSF rhinorrhea is the most common and debilitating complication of TSA. The incidence of postoperative CSF rhinorrhea after TSA has been reported to range from 0.5 to 6.0 %. During surgery, it is very important not to violate the arachnoid membrane to prevent this problem; however, it is often unavoidable. Moreover, the membrane can be violated and torn more frequently during the removal of a capsule because this technique requires more aggressive resection of a pseudocapsule from the pituitary gland and arachnoid membrane attached directly to the surface of pseudocapsule. Indeed, in our series, the chance of intraoperative CSF leakage was higher in patients with extracapsular resection than in those with a piecemeal resection. However, the risk of postoperative CSF rhinorrhea was only 4.2 % in in an extracapsular resection group. Although the chance of postoperative CSF rhinorrhea was a little higher than that of a piecemeal group, the difference
was not statistically significant. Interestingly, the incidence of postoperative CSF rhinorrhea was higher in the fragmented capsulectomy group than in the en bloc resection group. The main reason for this is that, in the former group, an additional maneuver was necessary to remove small pseudocapsule fragments from a thin normal gland and arachnoid membrane, which led to an increased risk of CSF leakage. In our early experience, we used conventional sellar floor repair methods using autologous tissue grafts such as fat, muscle, or tensor fascia lata with or without postoperative lumbar drainage of CSF. However, since 2005, we tried to repair the primary leaking point directly with various techniques according to the type of intraoperative leakage: collagen fleece coated with a fibrin sealant (TachoComb; Nycomed Austria, Linz, Austria) [13]; an arachnoid membrane coaptation technique [13]; direct suture of the torn arachnoid membrane; and a fascia repair technique. Although the removal of a pseudocapsule increases the risk of intraoperative CSF leakage, we now believe it can be managed by proper closure techniques without the aid of postoperative lumbar CSF drainage. Hypopituitarism Another concern regarding the possible drawbacks of capsulectomy is that removal of a pseudocapsule might increase the risk of postoperative hypopituitarism. Qu et al. [6] emphasized that the extracapsular resection technique helps to identify and discriminate pituitary adenomas from the normal pituitary gland so that it reduces the risk of erroneous removal of a normal pituitary gland [14]. However, it is also true that additional removal of a pseudocapsule is more likely to injure the remaining pituitary gland and consequently increases the risk of worsening of pituitary function. In our previous report [3], we demonstrated that removal of a pseudocapsule did not worsen postoperative pituitary function. In our present series, we also showed that the incidence of postoperative worsening of pituitary function was not different according to tumor resection technique. From a technical aspect, it is very important to find the correct dissection plane during capsulectomy. Sometimes, a thinned normal pituitary gland looks very similar to a pseudocapsule. If a surgeon misjudges this as a pseudocapsule and dissect outside the normal pituitary gland, the normal pituitary gland would be taken out together with the capsule, which would abolish normal pituitary function. Also, as described in the surgical procedure, we remove a pseudocapsule and very carefully protect the normal pituitary gland with Bemsheet. Simply pulling a pseudocapsule without this maneuver could violate the normal pituitary gland and increase the chance of postoperative hypopituitarism.
123
Pituitary
Conclusions Incomplete removal of a pseudocapsule often leads to failure of radical tumor resection and endocrinological remission. When a pseudocapsule is well developed and used as a surgical dissection plane, the tumor can be removed an in extracapsular fashion with the best outcome for the patient. When a pseudocapsule develops incompletely or is very faint or thin, extracapsular dissection should be attempted as far as possible and any suspicious pseudocapsule should be removed completely. The removal of a pseudocapsule does not increase the risk of postoperative hypopituitarism nor postoperative CSF rhinorrhea. However, violent surgical maneuvers should be avoided as these might injure the nearby normal pituitary gland and its stalk and lead to an increased risk of postoperative hypopituitarism. Acknowledgments The authors thank Juyoon Park, R.N., M.P.H., O.C.N., Min Kyeong Jang, R.N., Sung Ja Kang, R.N., and Bok Soon Lee, R.N., Pituitary Tumor Clinic and Soo Yeon Choi, M.P.H., Department of Medical Recording for their tremendous effort in performing the endocrinological tests and data acquisition for such a long follow-up duration. Conflict of interest The authors report no conflict of interest concerning the materials or methods used in this study and there was no financial support or relationship with any organization.
References 1. Costello RT (1936) Subclinical adenoma of the pituitary gland. Am J Pathol 12:205–216 2. Oldfield EH, Vortmeyer AO (2006) Development of a histological pseudocapsule and its use as a surgical capsule in the excision of pituitary tumors. J Neurosurg 104:7–19
123
3. Lee EJ, Ahn JY, Noh T, Kim SH, Kim TS (2009) Tumor tissue identification in the pseudocapsule of pituitary adenoma: should the pseudocapsule be removed for total resection of pituitary adenoma? Neurosurgery 64:ons62–ons69 4. Jagannathan J, Smith R, DeVroom HL, Vortmeyer AO, Stratakis CA, Nieman LK, Oldfield EH (2009) Outcome of using the histological pseudocapsule as a surgical capsule in Cushing disease. J Neurosurg 111:531–539 5. Kawamata T, Kubo O, Hori T (2005) Surgical removal of growth hormone-secreting pituitary adenomas with intensive microsurgical pseudocapsule resection results in complete remission of acromegaly. Neurosurg Rev 28:201–208 6. Qu X, Yang J, Sun J, Mou C, Wang G, Han T, Qu YM, Wang M, Xu GM (2011) Transsphenoidal pseudocapsule-based extracapsular resection for pituitary adenomas. Acta Neurochir (Wien) 153:799–806 7. Ceylan S, Cabuk B, Koc K, Anik I, Vural C (2013) Endoscopic distinction between capsule and pseudocapsule of pituitary adenomas. Acta Neurochir (Wien) 155:1611–1619 8. Prevedello D, Ebner F, Lara D, Filho L, Otto B, Carrau R (2013) Extracapsular dissection technique with the Cotton Swab for pituitary adenomas through an endoscopic endonasal approach— how I do it. Acta Neurochir (Wien) 155:1629–1632 9. Ku CR, Kim EH, Oh MC, Lee EJ, Kim SH (2012) Surgical and endocrinological outcomes in the treatment of growth hormonesecreting pituitary adenomas according to the shift of surgical paradigm. Neurosurgery 71:ons192–ons203 10. Oh MC, Kim EH, Kim SH (2012) Coexistence of intracranial aneurysm in 800 patients with surgically confirmed pituitary adenoma. J Neurosurg 116:942–947 11. Monteith SJ, Starke RM, Jane JA, Oldfield EH (2012) Use of the histological pseudocapsule in surgery for Cushing disease: rapid postoperative cortisol decline predicting complete tumor resection. J Neurosurg 116:721–727 12. Laws ER (2006) Pituitary pseudocapsule. J Neurosurg 104:1–2 13. Cho JM, Ahn JY, Chang JH, Kim SH (2011) Prevention of cerebrospinal fluid rhinorrhea after transsphenoidal surgery by collagen fleece coated with fibrin sealant without autologous tissue graft or postoperative lumbar drainage. Neurosurgery 68:130–136 14. Qu X, Xu G, Qu Y, Song T (2011) The pseudocapsule surrounding a pituitary adenoma and its clinical significance. J Neurooncol 101:171–178
Extracapsular en bloc resection in pituitary adenoma surgery Eui Hyun Kim • Cheol Ryong Ku • Eun Jig Lee Sun Ho Kim
•
Ó Springer Science+Business Media New York 2014
Abstract Purpose A pituitary pseudocapsule often contains tumor tissue and should be removed for radical resection. It can be used as a surgical plane for more radical resection of the tumor in many cases of pituitary adenomas. We evaluated the advantages and disadvantages of extracapsular en bloc capsulectomy. Methods From 1992 until 2011, 1,089 treated patients were grouped according to the resection technique: en bloc capsulectomy, fragmented capsulectomy, or piecemeal resection. Their surgical and endocrinological outcomes and complications were evaluated. Results Extracapsular tumor resection was performed in 263 patients; en bloc capsulectomy in 94 patients and
Electronic supplementary material The online version of this article (doi:10.1007/s11102-014-0587-4) contains supplementary material, which is available to authorized users. E. H. Kim S. H. Kim (&) Neurosurgery, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea e-mail: [email protected] E. H. Kim C. R. Ku E. J. Lee S. H. Kim Endocrine Research Institute, Yonsei University College of Medicine, Seoul, Republic of Korea E. H. Kim C. R. Ku E. J. Lee S. H. Kim Pituitary Tumor Clinic, Yonsei University College of Medicine, Seoul, Republic of Korea C. R. Ku E. J. Lee Endocrinology, Yonsei University College of Medicine, Seoul, Republic of Korea E. J. Lee Endocrinology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA
fragmented capsulectomy in 169, whereas piecemeal resection was performed in 826. Extracapsular resection was performed more frequently in prolactin- and thyroidstimulating hormone-secreting tumors. Total resection was more frequently achieved in extracapsular resection and its chance was 100 % when tumors were removed in an en bloc fashion. For the functioning pituitary adenomas, endocrinological remission was achieved in all patients whose tumors were removed in an en bloc fashion and there was no recurrence. Postoperative cerebrospinal fluid (CSF) rhinorrhea developed in 4.2 and 2.7 % in the extracapsular resection group and the piecemeal resection groups, respectively. The chance of postoperative aggravation of pituitary function was not statistically different between groups. Conclusions Extracapsular resection is critical for radical tumor resection and endocrinological remission. The removal of a pseudocapsule does not increase the risk of postoperative hypopituitarism nor postoperative CSF rhinorrhea. Keywords Capsulectomy En bloc resection Extracapsular resection Pituitary adenoma Pseudocapsule
Introduction Since a pseudocapsule associated with a pituitary adenoma was initially described in the 1900s [1], there have been many reports about pituitary pseudocapsules. These are thought to develop in response to pressure within the pituitary gland itself and become thin multiple layers of compressed pituitary gland tissue containing a reticulin network. It is found in about 50 % of patients with a
123
Pituitary
pituitary adenoma and tends to be more frequent in larger tumors [2]. Histopathology has revealed that tumor tissue is frequently present within the pseudocapsule, which justifies the aggressive resection especially in endocrine-active pituitary tumors [3]. Thus, in recent years, many authors have adopted so-called extracapsular resection and emphasized the importance of a pituitary pseudocapsule as a surgical plane for more radical resection of the tumor [2– 8]. However, most authors have focused on the surgical technique and histological findings rather than on surgical and endocrinological outcomes. It has been our surgical strategy to perform intensive resection of pseudocapsules as we also found they could be a source of tumor recurrence and failure of remission in endocrine-active tumors [3]. Tumor dissection was performed outside the pseudocapsule whenever possible. When extracapsular resection was not possible, tumors were removed in a piecemeal fashion. We grouped our patients based on the resection techniques and evaluated surgical and endocrinological outcomes and complications. The advantages and disadvantages of extracapsular resection are discussed below with a description of our surgical techniques.
Methods Patients From January 1992 until December 2011, 1,372 cases of pituitary adenoma were operated on with a transsphenoidal approach (TSA) at Yonsei University College of Medicine by a single neurosurgeon (SHK). All adenomas were classified according to a modified Hardy radiological classification scheme [9, 10]. Of these, 283 patients were excluded from the study for the following reasons: Hardy type IV tumors, which had cavernous sinus invasion (n = 210); revision surgery for recurred tumors (n = 67); history of radiotherapy (n = 2); and loss to follow-up (n = 4). Thus, 1,089 patients were included. These included 464 male and 625 female patients with a mean age of 43.4 years. The follow-up duration ranged from 1.0 to 18.4 years (mean 4.8 years). The classification of pituitary adenomas according to hormonal activity was verified from postoperative immunohistochemistry of the tumor tissue. Endocrine-active tumors were observed in 537 patients (49.3 %); growth hormone (GH)-secreting tumors were observed in 312, prolactin (PRL)-secreting tumors were observed in 163, adrenocorticotropic hormone (ACTH)secreting tumors were observed in 39, thyroid-stimulating hormone (TSH)-secreting tumors were observed in 15 and both GH- and PRL-secreting tumors were observed in 8 patients. The presence of nonfunctioning (NF) tumors was
123
confirmed in 552 patients (50.7 %). Medical treatment with a dopamine agonist was tried prior to surgery in 80 patients (49.1 %) with PRL-secreting tumors and they showed no or insufficient response. Among the patients with GHsecreting tumors, 19 patients had ever had preoperative octreotide long-acting release (Sandostatin LAR Depot; Novartis Pharmaceuticals Corp., East Hanover, NJ, USA) treatment. For 38 patients with NF pituitary adenomas, medical treatment was tried under the impression of a PRLsecreting tumor, but this failed prior to the surgical treatment. Patient evaluation The preoperative patient work-up included taking a complete history, physical and neurological examinations, and endocrinological and radiological assessments. All patients underwent preoperative dynamic magnetic resonance imaging (MRI) of the sellar region using a 1.5-T system (Signa; General Electric, Minneapolis, MN, USA) or a 3.0-T system (Achieva; Philips, Best, The Netherlands). Immediate postoperative 3-T diagnostic MRI was performed within 48 h after the surgery and the radicality of tumor removal was determined. Follow-up MRI was performed 1 year after surgery and every other year thereafter. The endocrinological assessment with combined pituitary function test (CPFT) was fully described in our previous report [3]. The changes in pituitary function from before and after TSA were classified into four categories: (1) normal to normal, meaning no pituitary hormone deficiency before and after TSA; (2) improved hypopituitarism, meaning the hypopituitarism improved including complete recovery; (3) persistent hypopituitarism, meaning the degree of pituitary hormone deficiency did not change after TSA; and (4) worsened hypopituitarism, meaning the number and/or degree of pituitary hormone deficiency worsened after TSA. The last of these categories included the new development of postoperative hypopituitarism and the aggravation of preoperatively existing hypopituitarism. CPFT was performed to evaluate anterior pituitary function before surgery, 6–12 months after surgery, and every 2 years thereafter. In acromegalic patients, an oral glucose tolerance test (OGTT, 75 g glucose) was performed preoperatively, 1 week after surgery, every 6 months up to 3 years after surgery, and then annually. Serum insulin-like growth factor-1 (IGF-1) was also measured whenever OGTT was performed. Remission was defined when the nadir serum GH level after an OGTT was \1 lg/L with normalization of serum IGF-1 adjusted for age and sex. Patients with Cushing’s disease were evaluated by measurement of 24 h urinary free cortisol and 8 AM serum cortisol level, serum ACTH level, and dexamethasone suppression test, for preoperative diagnosis and determination of remission. The levels of PRL, T3, T4, and
Pituitary
TSH were evaluated in patients with PRL- and TSHsecreting tumors, respectively.
(version 20.0; IBM, Armonk, NY, USA) and P \ .05 was taken as significant.
Surgical procedure Results The procedure was performed in the same way as in conventional microscopic TSA. Tumors were approached differently according to their location and size. When a tumor was not fully visualized or when a remnant tumor was suspicious on an invisible area under the microscope, an endoscope was introduced into the surgical field. After dural opening, we localized the normal pituitary gland first and checked for the consistency of the tumor and the presence of a pseudocapsule. If a tumor showed hard consistency or had a fairly thick pseudocapsule, extracapsular dissection was performed outside it. In this situation, special caution was taken not to dissect outside a normal pituitary gland with improper orientation. If a pseudocapsule covered the entire surface of a tumor, en bloc resection was attempted (en bloc capsulectomy) (Video 1–4). When pseudocapsule formation was incomplete, the tumor was removed as several fragmented pieces (fragmented capsulectomy). When a tumor showed fragile and soft consistency with a scarcely developed pseudocapsule, the tumor was removed in a piecemeal fashion (piecemeal resection). In these cases, any pseudocapsule found over the normal pituitary gland surface was always removed completely and this suspicious tissue was sent to the pathology laboratory for frozen section histopathology. With any faint suspicion of dural invasion, all exposed dura mater was removed as much as possible. When a pseudocapsule was attached onto the overlying thin layer of the normal gland and arachnoid membrane, these were protected with a small piece of Bemsheet (Kawamoto Bandage Material Co., Ltd, Osaka, Japan) and the pseudocapsule was separated very carefully. Even when a pseudocapsule was very adhesive to the arachnoid membrane, gentle manipulation was always mandatory and a small piece of Bemsheet was very useful during this step. Rough pulling of the pseudocapsule was never attempted. Even though dissection was performed very carefully, sometimes cerebrospinal fluid (CSF) leakage encountered. When this was noted during surgery, it was sealed by various direct closure techniques to close the leaking point. Postoperative lumbar drainage of CSF was performed in many of our early cases; however, it has not been used since 2005. Statistical analysis Chi squared and Fisher’s exact tests were used for determining the statistical significance of any differences. Statistical analyses were performed using IBM SPSS Statistics
During surgery, a pseudocapsule was identified in 531 patients (48.8 %). Among these patients, a pseudocapsule was apparent and thick enough to be used as a surgical dissection plane in only 263 (49.5 %). En bloc capsulectomy was achieved in 94 patients and fragmented capsulectomy in 169. The tumors were removed in a piecemeal fashion in 826 patients (75.8 %): 268 with a fragile, fragmented pseudocapsule and 558 with no visible pseudocapsule. Tumor removal fashion according to the type of hormone produced is summarized in Table 1. When compared with the manner of tumor removal in all patients, extracapsular resection was performed more frequently in PRL-secreting tumors and in TSH-secreting tumors whereas NF pituitary adenomas were more likely to be removed in a piecemeal fashion. These differences were statistically significant. Extent of resection The results for the extent of resection are summarized in Table 2. Tumors were removed totally in 258 patients (98.1 %) among 263 patients in whom extracapsular resection was possible. When tumors were removed in an en bloc fashion, total resection was achieved in all 94 patients. In the group with fragmented capsulectomy, total resection was possible in 164 of 169 patients (97.0 %). However, when tumors were removed in a piecemeal fashion, only 93.6 % of patients (773/826) achieved total resection. These differences were statistically significant (P = .004). Unexpected remnant tumors were found on the postoperative MRI in 6 of 826 patients (0.7 %) with a piecemeal resection whereas no patients with extracapsular Table 1 Tumor removal fashion according to the hormonal type of tumor Hormonal type
Extracapsular resection
Piecemeal resection
P value
GH
80 (25.6 %)
232 (74.4 %)
.466
PRL
56 (34.4 %)
107 (65.6 %)
.001
TSH
\.001
11 (73.3 %)
4 (26.7 %)
ACTH
4 (10.3 %)
35 (89.7 %)
.037
GH ? PRL
3 (37.5 %)
5 (62.5 %)
.409
NF
109 (19.7 %)
443 (80.3 %)
.001
Total
263 (24.2 %)
826 (75.8 %)
\.001
ACTH adrenocorticotropin hormone, GH growth hormone, NF nonfunctioning, PRL prolactin, TSH thyroid stimulating hormone
123
Pituitary Table 2 Radicality of tumor removal according to tumor removal fashion
Extracapsular resection En bloc capsulectomy Total resection
98.1 % (258/263) 100 % (94/94)
a
Unexpected remnant tumors
0 % (0/263)
Recurrence
3.1 % (8/258a)
0 % (0/94)
Number of patients with total resection
0 % (0/94 )
Table 3 Endocrinological remission rate according to the hormonal type of the tumor
Extracapsular resection
a
Total number of patients with functioning tumors whose biochemical remission was determined postoperatively
a
P value
93.6 % (773/826)
.004
0.7 % (6/826)
.345
1.3 % (10/773a)
.055
97.0 % (164/169) 0 % (0/169) a
4.9 % (8/164 )
Piecemeal resection
Total
P value
GH
97.4 % (75/77)
88.5 % (193/218)
90.8 % (268/295)
.020
PRL
77.6 % (38/49)
73.1 % (68/93)
74.6 % (106/142)
.564
TSH
100 % (11/11)
75.0 % (3/4)
93.3 % (14/15)
.267
ACTH GH ? PRL
100 % (4/4) 50 % (1/2)
100 % (35/35) 75.0 % (3/4)
100 % (39/39) 66.7 % (4/6)
– 1.000
Total
90.2 % (129/143)
85.3 % (302/354)
86.7 % (431/497a)
.145
En bloc capsulectomy
Others
Total
P value
GH
100 % (29/29)
89.8 % (239/266)
90.8 % (268/295)
.089
PRL
100 % (12/12)
72.3 % (94/130)
74.6 % (106/142)
.037
TSH
100 % (8/8)
85.7 % (6/7)
93.3 % (14/15)
.467
ACTH
100 % (2/2)
GH ? PRL Total
0% 100 % (51/51)
resection revealed any unexpected remnant tumors on their immediate postoperative MRI. However, the difference was not statistically significant (P = .345). When their tumors were totally removed, the recurrence rate 3.1 % (8/ 258) in the group with extracapsular resection, which was a little higher than that of a piecemeal resection group (1.3 %, 10/773) even though it was not statistically significant (P = .055). However, when the tumors were removed in en bloc resection technique, the recurrence rate was zero. Biochemical remission in endocrine-active tumors Postoperative determination of biochemical remission was feasible in 497 of 537 patients with endocrine-active tumors as 40 patients were lost to follow-up. Overall, 431 patients (86.7 %) showed biochemical remission (Table 3). The remission chance was a little higher in an extracapsular resection group (90.2 %, 129/143) than in a piecemeal resection group (85.3 %, 302/354), however, the difference was not statistically significant. When we grouped the patients into an en bloc capsulectomy group and others, as 51 patients with en bloc resection showed a 100 % biochemical remission rate, the patients with en bloc resection
123
Fragmented capsulectomy
Piecemeal resection
100 % (37/37)
100 % (39/39)
–
66.7 % (4/6)
66.7 % (4/6)
–
85.2 % (380/446)
86.7 % (431/497a)
.001
Table 4 Influence of preoperative medical treatment on the tumor removal fashion Extracapsular resection Total number of patients with preoperative medication (n = 137)
49 (35.5 %)
Piecemeal resection 89 (64.5 %)
P value .001
GH (n = 19)
4 (21.1 %)
15 (78.9 %)
.750
PRL (n = 80)
32 (40.0 %)
48 (60.0 %)
.001
NF (n = 38)
12 (31.6 %)
26 (68.4 %)
.276
94 (8.6 %)
826 (75.9 %)
Total number of patients (n = 1,089)
GH growth hormone, NF non-functioning, PRL prolactin
showed much higher chance of biochemical cure than other patients with statistical significance (P = .001). Preoperative medical treatment was attempted in 80 patients with PRL-secreting pituitary adenomas, 19 patients with GH-secreting pituitary adenomas, and 38 NF pituitary adenomas; insufficient responses were documented in all of them before surgical treatment (Table 4). When these 137 patients were classified based on the tumor removal fashion, the patients with preoperative medical
Pituitary Table 5 Incidence of CSF leak according to tumor removal fashion
Intraoperative CSF leak
Extracapsular resection
Piecemeal resection
P value
59.3 % (156/263)
41.6 % (344/826)
\.001
Extracapsular resection
Postoperative CSF rhinorrhea
En bloc resection
Fragmented capsulectomy
4.2 % (11/263) 1.1 % (1/94)
5.9 % (10/169)
Piecemeal resection
P value
2.7 % (22/826)
.211
CSF cerebrospinal fluid
Table 6 Postoperative hormonal outcome according to tumor removal fashion Postoperative hormonal outcome
Extracapsular resection
Piecemeal resection 137 (19.0 %)
Preserved normal pituitary function (n = 189, 19.7 %)
52 (22.1 %)
Improved hypopituitarism (n = 505, 52.7 %)
119 (50.6 %)
386 (53.4 %)
Persistent hypopituitarism (n = 146, 15.3 %)
28 (12.0 %)
118 (16.3 %)
Aggravation of hypopituitarism (n = 118, 12.3 %)
36 (15.3 %)
82 (11.3 %)
235
723
Total (n = 958) a
a
P value
.127
Total number of patients in whom comparison between preoperative and postoperative CPFT was available
treatment were more likely to undergo extracapsular resection (P = .001) compared with tumor removal fashion in total patients. However, when patients were subclassified into each hormonal type of tumors, this difference was observed only in patients with PRL-secreting pituitary adenomas (P = .001).
(12.3 %). When they were classified based on the fashion of tumor removal, no statistically significant differences were found (P = .127).
Cerebrospinal fluid leak
The presence of pseudocapsules was first described in the early 1900s; however, understanding their formation and highlighting their clinical significance have been recent developments. Oldfield and Vortmeyer [2] described how histological pseudocapsule develops by examining the distribution of tumor size in relation to the presence of a pseudocapsule. They emphasized that the pseudocapsule is extremely useful for complete tumor excision. There have been many reports describing surgical techniques using a pseudocapsule as a surgical dissection plane. In those reports, the dissection was performed along the interface between the outer surface of a pseudocapsule and a normal pituitary gland [3–8]. Their techniques increased the complete tumor removal and endocrinological remission rates and decreased the complication and recurrence rates. Qu et al. [6] defined these techniques as pseudocapsulebased extracapsular resection approach. However, this technique is not always possible because whether the resection is performed outside or inside the pseudocapsule largely depends on the consistency of the tumor and the degree of development of the pseudocapsule. In our early experience, we demonstrated that a microsurgical pseudocapsule was found in 55.7 % of patients
Intraoperative CSF leakage was noted in 156/263 (59.3 %) of patients with extracapsular resection and 344/826 (41.6 %) in the group with piecemeal resection (Table 5). The difference in the risks of intraoperative CSF leakage were statistically significant with P values of \.001. Postoperative CSF rhinorrhea developed in 11/263 (4.2 %) of patients with extracapsular resection, which was a little higher than the 22/826 (2.7 %) among patients with piecemeal resection. However, the difference in the risks of postoperative CSF leakage were statistically not significant (P = .211). The possibility of postoperative CSF rhinorrhea was higher in a fragmented capsulectomy group (5.9 %) than in an en bloc resection group (1.1 %) and in piecemeal resection group (2.7 %). Hypopituitarism Comparisons between preoperative and postoperative CPFT were available in 958 patients (Table 6). Overall, normal pituitary function was preserved in 189 patients (19.7 %), improved in 505 (52.7 %), and aggravated in 118
Discussion
123
Pituitary
Fig. 1 Photomicrographs of tumor specimens. a Tumor specimen obtained en bloc with a distinct pseudocapsule during surgery, showing a relatively thick layer of connective tissue at the interface between the pituitary adenoma and the normal pituitary gland;
hematoxylin and eosin (H & E) staining, original magnification 940. b With higher magnification, clusters of tumor cells can be identified in the pseudocapsule; H & E staining, original magnification 9200
with pituitary adenomas [3]. The degree and characteristics of pseudocapsule varied widely: (1) entirely covering the tumor mass; (2) a thin, fibrous envelope; (3) a yellowish, discolored, normal gland-like, thin membrane; (4) thick, fibrous tissue; and (5) sometimes dense fibrosis or calcifications. We believe the tumor size, hormonal type, and the presence of intratumoral hemorrhage can influence the nature of the pseudocapsule. Thus, different surgical techniques should be adopted according to the presence of a surgically resectable pseudocapsule. When a tumor has a fibrous and thick pseudocapsule or when a tumor itself has very hard and fibrous consistency, dissection should be performed outside the pseudocapsule (Fig. 1). By contrast, if the tumor is very soft and fragile and its pseudocapsule is very thin or faint, piecemeal resection is inevitable. However, even in the latter situation, close inspection is essential to check for the presence of any suspicious pseudocapsule over the remaining pituitary gland or arachnoid membrane and this should be removed if it is found.
recurrence, although it was not statistically significant, the recurrence rate was slightly higher in an extracapsular resection group than in a piecemeal resection group, which was mainly because we experienced more recurrent tumors in a group of fragmented capsulectomy. Considering no patients experienced recurrence in the group with en bloc resection, we presume the chance of a small fragment of residue might be higher in cases of a fragmented capsulectomy because fibrosis, calcification, and incompleteness of pseudocapsule were more frequent in this group of patients. Therefore, careful inspection is essential with the suspicion that tumor cells might still reside beyond the main tumor boundary when a pseudocapsule is incompletely developed.
Extent of resection Pseudocapsule-based extracapsular resection is a technique that maximizes the extent of resection, as the tumor is resected completely within its pseudocapsule. The pseudocapsule itself can be used as a reliable boundary of the inner tumor content. In our series, the rate of total resection was much higher in the groups with en bloc resection and fragmented capsulectomy than in the piecemeal group. Although there was no statistical difference in the incidence of unexpected remnants among the three groups (P = .383), no patients who underwent en bloc resection or fragmented capsulectomy showed remnant tumors on their postoperative MRIs. As for
123
Biochemical remission in endocrine-active tumors An increased extent of resection leads to increased remission rates among patients with endocrine-active pituitary adenomas, as demonstrated for GH-secreting pituitary adenomas by Ku et al. [9] and Kawamata et al. [5], and for ACTH-secreting pituitary adenomas by Jagannathan et al. [4] and Monteith et al. [11]. Although this technique is only possible when the tumor has quite a well-developed pseudocapsule, the authors emphasized that the capsule itself contains tumor cells and can be a main cause of persistent hypersecretion of the hormone and possibly the source of recurrence. Also, in our previous reports, we suggested that complete removal of the pseudocapsule is critical to achieve endocrinological remission in patients with endocrine-active pituitary adenomas [3, 9]. However, even in a tumor with a well-developed pseudocapsule, tumor cells might invade the nearby normal pituitary gland and dural layer beyond the pseudocapsule [12]. This is known to be one of the main reasons for discordance
Pituitary
between complete tumor removal and endocrinological remission. For this reason, Laws [12] emphasized that it is necessary to remove a rim of normal pituitary gland surrounding the tumor to improve surgical outcome in patients with endocrine-active pituitary adenomas. However, this fact does not diminish the usefulness of pseudocapsulebased extracapsular resection. This technique enables reliable complete resection of inner tumor contents. Moreover, the pseudocapsule itself might be infiltrated by tumor cells, which justifies the radical removal of the pseudocapsule. Our previous study showed that the presence of a pseudocapsule was more frequent in PRL-secreting pituitary adenomas than in GH- and ACTH-secreting pituitary adenomas [3]. Also, our present study demonstrated that extracapsular resection was more feasible only in PRLsecreting pituitary adenomas. We have already suggested that this might be associated with previous medical treatment with dopamine agonists, which facilitate fibrous tissue formation. Although preoperative medical treatment seems to enhance the formation of a pseudocapsule, it also makes the interface between pseudocapsule and a normal pituitary gland very adhesive, which in turn makes the surgical maneuver quite difficult. As discussed above, we verified that removal of the pseudocapsule increases the extent of resection and endocrinological remission rate. However, because this technique requires more aggressive resection of a pseudocapsule from the surface of an underlying pituitary gland and arachnoid membrane than the piecemeal resection technique, it could increase the possibility of complications. We also tried to clarify our concerns about complications, which are as important as the radicality of tumor removal. Cerebrospinal fluid leakage CSF rhinorrhea is the most common and debilitating complication of TSA. The incidence of postoperative CSF rhinorrhea after TSA has been reported to range from 0.5 to 6.0 %. During surgery, it is very important not to violate the arachnoid membrane to prevent this problem; however, it is often unavoidable. Moreover, the membrane can be violated and torn more frequently during the removal of a capsule because this technique requires more aggressive resection of a pseudocapsule from the pituitary gland and arachnoid membrane attached directly to the surface of pseudocapsule. Indeed, in our series, the chance of intraoperative CSF leakage was higher in patients with extracapsular resection than in those with a piecemeal resection. However, the risk of postoperative CSF rhinorrhea was only 4.2 % in in an extracapsular resection group. Although the chance of postoperative CSF rhinorrhea was a little higher than that of a piecemeal group, the difference
was not statistically significant. Interestingly, the incidence of postoperative CSF rhinorrhea was higher in the fragmented capsulectomy group than in the en bloc resection group. The main reason for this is that, in the former group, an additional maneuver was necessary to remove small pseudocapsule fragments from a thin normal gland and arachnoid membrane, which led to an increased risk of CSF leakage. In our early experience, we used conventional sellar floor repair methods using autologous tissue grafts such as fat, muscle, or tensor fascia lata with or without postoperative lumbar drainage of CSF. However, since 2005, we tried to repair the primary leaking point directly with various techniques according to the type of intraoperative leakage: collagen fleece coated with a fibrin sealant (TachoComb; Nycomed Austria, Linz, Austria) [13]; an arachnoid membrane coaptation technique [13]; direct suture of the torn arachnoid membrane; and a fascia repair technique. Although the removal of a pseudocapsule increases the risk of intraoperative CSF leakage, we now believe it can be managed by proper closure techniques without the aid of postoperative lumbar CSF drainage. Hypopituitarism Another concern regarding the possible drawbacks of capsulectomy is that removal of a pseudocapsule might increase the risk of postoperative hypopituitarism. Qu et al. [6] emphasized that the extracapsular resection technique helps to identify and discriminate pituitary adenomas from the normal pituitary gland so that it reduces the risk of erroneous removal of a normal pituitary gland [14]. However, it is also true that additional removal of a pseudocapsule is more likely to injure the remaining pituitary gland and consequently increases the risk of worsening of pituitary function. In our previous report [3], we demonstrated that removal of a pseudocapsule did not worsen postoperative pituitary function. In our present series, we also showed that the incidence of postoperative worsening of pituitary function was not different according to tumor resection technique. From a technical aspect, it is very important to find the correct dissection plane during capsulectomy. Sometimes, a thinned normal pituitary gland looks very similar to a pseudocapsule. If a surgeon misjudges this as a pseudocapsule and dissect outside the normal pituitary gland, the normal pituitary gland would be taken out together with the capsule, which would abolish normal pituitary function. Also, as described in the surgical procedure, we remove a pseudocapsule and very carefully protect the normal pituitary gland with Bemsheet. Simply pulling a pseudocapsule without this maneuver could violate the normal pituitary gland and increase the chance of postoperative hypopituitarism.
123
Pituitary
Conclusions Incomplete removal of a pseudocapsule often leads to failure of radical tumor resection and endocrinological remission. When a pseudocapsule is well developed and used as a surgical dissection plane, the tumor can be removed an in extracapsular fashion with the best outcome for the patient. When a pseudocapsule develops incompletely or is very faint or thin, extracapsular dissection should be attempted as far as possible and any suspicious pseudocapsule should be removed completely. The removal of a pseudocapsule does not increase the risk of postoperative hypopituitarism nor postoperative CSF rhinorrhea. However, violent surgical maneuvers should be avoided as these might injure the nearby normal pituitary gland and its stalk and lead to an increased risk of postoperative hypopituitarism. Acknowledgments The authors thank Juyoon Park, R.N., M.P.H., O.C.N., Min Kyeong Jang, R.N., Sung Ja Kang, R.N., and Bok Soon Lee, R.N., Pituitary Tumor Clinic and Soo Yeon Choi, M.P.H., Department of Medical Recording for their tremendous effort in performing the endocrinological tests and data acquisition for such a long follow-up duration. Conflict of interest The authors report no conflict of interest concerning the materials or methods used in this study and there was no financial support or relationship with any organization.
References 1. Costello RT (1936) Subclinical adenoma of the pituitary gland. Am J Pathol 12:205–216 2. Oldfield EH, Vortmeyer AO (2006) Development of a histological pseudocapsule and its use as a surgical capsule in the excision of pituitary tumors. J Neurosurg 104:7–19
123
3. Lee EJ, Ahn JY, Noh T, Kim SH, Kim TS (2009) Tumor tissue identification in the pseudocapsule of pituitary adenoma: should the pseudocapsule be removed for total resection of pituitary adenoma? Neurosurgery 64:ons62–ons69 4. Jagannathan J, Smith R, DeVroom HL, Vortmeyer AO, Stratakis CA, Nieman LK, Oldfield EH (2009) Outcome of using the histological pseudocapsule as a surgical capsule in Cushing disease. J Neurosurg 111:531–539 5. Kawamata T, Kubo O, Hori T (2005) Surgical removal of growth hormone-secreting pituitary adenomas with intensive microsurgical pseudocapsule resection results in complete remission of acromegaly. Neurosurg Rev 28:201–208 6. Qu X, Yang J, Sun J, Mou C, Wang G, Han T, Qu YM, Wang M, Xu GM (2011) Transsphenoidal pseudocapsule-based extracapsular resection for pituitary adenomas. Acta Neurochir (Wien) 153:799–806 7. Ceylan S, Cabuk B, Koc K, Anik I, Vural C (2013) Endoscopic distinction between capsule and pseudocapsule of pituitary adenomas. Acta Neurochir (Wien) 155:1611–1619 8. Prevedello D, Ebner F, Lara D, Filho L, Otto B, Carrau R (2013) Extracapsular dissection technique with the Cotton Swab for pituitary adenomas through an endoscopic endonasal approach— how I do it. Acta Neurochir (Wien) 155:1629–1632 9. Ku CR, Kim EH, Oh MC, Lee EJ, Kim SH (2012) Surgical and endocrinological outcomes in the treatment of growth hormonesecreting pituitary adenomas according to the shift of surgical paradigm. Neurosurgery 71:ons192–ons203 10. Oh MC, Kim EH, Kim SH (2012) Coexistence of intracranial aneurysm in 800 patients with surgically confirmed pituitary adenoma. J Neurosurg 116:942–947 11. Monteith SJ, Starke RM, Jane JA, Oldfield EH (2012) Use of the histological pseudocapsule in surgery for Cushing disease: rapid postoperative cortisol decline predicting complete tumor resection. J Neurosurg 116:721–727 12. Laws ER (2006) Pituitary pseudocapsule. J Neurosurg 104:1–2 13. Cho JM, Ahn JY, Chang JH, Kim SH (2011) Prevention of cerebrospinal fluid rhinorrhea after transsphenoidal surgery by collagen fleece coated with fibrin sealant without autologous tissue graft or postoperative lumbar drainage. Neurosurgery 68:130–136 14. Qu X, Xu G, Qu Y, Song T (2011) The pseudocapsule surrounding a pituitary adenoma and its clinical significance. J Neurooncol 101:171–178
