Introduction

Transsphenoidal microscopic pituitary surgery has long been considered the gold standard in surgical treatment of pituitary tumors. Endonasal endoscopic pituitary surgery has come into prominence over the last two decades as an alternative to microscopic surgery. Each technique has its advantages and disadvantages; yet, for multiple reasons, a worldwide shift towards endonasal endoscopic surgery is in progress [1–5]. In recent years, multiple teams have published their experience with this transition [1, 2, 4, 6], and multiple systematic reviews and meta-analyses have been conducted on the comparison of these two techniques [7–13]. Though sometimes discordant on some issues, a global consensus favoring endoscopic surgery in most circumstances is slowly being reached. In this review, we use recent literature to discuss the advantages and pitfalls of both techniques.

Technically, the microscopic approach can be realized in two different fashions, through a sublabial incision or via an endonasal transseptal approach. Both routes involve the use of speculums to fracture the nasal septum and retract soft tissue. A sphenoidotomy is then performed using a self-retaining speculum that initially straddles the bony midline keel and retracts the nasal mucosa laterally [14]. Some extended microscopic transsphenoidal approaches have also been developed and have reported satisfying results [15–17]. Multiple nuances are also possible with endoscopic surgery. Depending on the case, the procedure can be performed through one or both nostrils following a posterior septectomy. While some authors use the endoscope holder at a certain stage of the surgery, others prefer a four handed technique [14], with two surgeons operating simultaneously.

Discussion

General principles of microscopic versus endoscopic transsphenoidal surgery

Both the microscope and endoscope have been used successfully to approach the sellar region and treat a wide variety of pathology encountered in this region. The large surgical series of microscopic approaches that emerged in the 1980s and 1990s defined surgical outcomes for pituitary adenomas, craniopharyngiomas, Rathke Cleft Cysts and a variety of additional pathology for decades to come [18–23]. Many of these outcomes still are considered the gold standard, and for many surgeons who are familiar with the microscope, it remains a viable tool with several major benefits. These benefits include maintaining stereoscopic vision, retraction of soft tissue that may at times be edematous or vascular, and in some cases a more rapid approach to the sella. For many neurosurgeons, using the microscope is more intuitive and does not require meeting the steep learning curve associated with endoscopic approaches. Finally, fully endoscopic approaches require multiple critical technological components that have to all be working properly, including the endoscope itself, camera, fiberoptic light cable, and video tower.

For a majority of surgeons who have been able to make the transition to fully endoscopic endonasal surgery, there is little doubt that this technique offers several key advantages that, upon surpassing the learning curve associated with endoscopic skull base surgery, make this technique clearly advantageous. These include a wider, more panoramic field of visualization, improved illumination and mobility of instruments, and an ability to look around anatomical corners using angled lenses. In addition, the ability to safely target regions of the skull base outside of the sella have paved the way for extended transcribriform, transplanum, transtuberculum, and transclival approaches, among others. High definition 3D endoscopic system usage is on the rise, which resolves the stereoscopic visualization issue [24].

The endoscopic approach offers two separate, but complementary, technical improvements over the microscope that improves anatomic visualization. The first involves the bony opening. The endoscopic approach requires a larger sphenoidotomy to permit the endoscope to be adequately positioned in the sinus along with two working instruments. This allows the endoscope to be moved around the sinus cavity to visualize larger areas of the skull base. In a complementary fashion, the endoscope lens provides a wider panoramic view, which is not confined by a direct linear viewing trajectory through a speculum as is the microscopic view. Moreover, endoscopes can have angled tips which can provide lateral exposure around anatomical corners that are not possible with the microscope [25, 26].

The microscope, on the other hand, provides stereoscopic vision not available with the endoscope, unless a 3D endoscopic system is used [24]. Concerns about instrument maneuverability during endoscopic procedures have largely been alleviated with the development of newer endoscope-specific instrumentation and the evolution of endoscopic techniques. These generally involve a requirement for instruments with low-profile, straight handles and mobile or curved tips to reach areas visualized using the endoscope. The requirement for bayoneted instruments used during microscopic approaches is obviated by a transition to fully endoscopic approaches. A recent report by Elhadi et al. provided quantitative data on the superiority of the binarial endoscopic approach in achieving target surgical freedom and sagittal angular freedom [14]. Varying degrees of “hybrid” endoscope-assisted microscopic surgery has been reported in the literature, with a unanimous observation that the endoscope not only allows better visualization of the sellar region, but also frequently helps in identifying small areas of residual tumor [27–31]. Improved visualization augments the confidence of the surgeon to be aggressive in resection of more fibrous tumors.

Surgical results

Gross total removal (GTR)

Studies comparing the extent of resection between endoscopic and microscopic surgery require very large numbers of patients to reach statistical significance since the results of microscopic surgery are already quite good. Furthermore, no class I data derived from direct comparisons of microscopic versus endoscopic approaches in randomized controlled trials is available, and may not be feasible. Therefore, surgeons have had to rely on comparisons of historical data to draw conclusions pertaining to the effectiveness and possible superiority of one technique over another, which have been classically skewed by selection bias and comparisons of surgical outcomes series that are decades apart. Nevertheless, the majority of meta-analysis and retrospective series converge on the slight superiority of endoscopic techniques in achieving gross total removal in pituitary tumors, especially when these tumors are locally invasive or are not just limited to the sella [1, 4, 7–10, 12, 13, 32].

In a recent study by Dallapiaza et al., tumors with Knosp grades 0–2 (i.e. those without cavernous sinus extension) underwent a similar extent of resection when approached via either approach [33]. However, in another paper by Messerer et al., a statistical difference emerged when comparing Knosp grade 2 and 3 patients. In the microscopy group, the GTR percentages for Knosp grade 2 and 3 patients were 47.8 and 16.7 % respectively, whereas endoscopy achieved much higher GTR rates of 88 and 69.9 % [32].

Significant suprasellar extension limits GTR rates achieved by both microsurgical [20] and endoscopic techniques [34]. However, when the height of the adenoma surpasses 30 mm, current data supports the superiority of endoscopic surgery to microscopic resection [20, 32, 35]. These results are likely associated with better visualization provided by the endoscope within the deeper operative blind spots beyond the edges of the bone work.

Overall, it is safe to say that the majority of pituitary and skull base surgeons find superiority in the endoscopic approach on account of its advantages in visualization, illumination, access, and mobility, as supported by the worldwide transition to the endoscopic approach.

Endocrinologic cure

The continuous evolution of the endocrinologic remission consensus criteria complicates any attempt to compare old historic microscopic results with those of more recent endoscopic patient series. The superiority of endoscopic surgery is still under investigation, but some early evidence already suggests better endocrinologic cure rates associated with endoscopic surgery. [4, 36] For functional adenomas, Razak et al. reported a remission rate of 57 % in the microscopic group and 94 % in the endoscopic group (p =  0.018). [4] Nevertheless, microsurgical results for surgical cure of microadenomas causing Cushing’s disease in highly experienced hands are outstanding and have not yet been matched by endoscopic series [37–39]. With respect to Cushing’s disease, it can be argued that one of the greatest advancement has been the pseudocapsular resection technique. To achieve this technique, Oldfield [37, 38] utilized a sublabial microscopic approach specifically to gain a wide surgical corridor, and more panoramic visualization of the sella. However, there is nothing specific about the microscopic approach that lends itself more to this technique. As more endoscopic surgeons adopt this technique the results should be comparable.

Visual outcome

Pre-operative visual disturbance is a frequent symptom and sign in patients with pituitary adenomas, particularly in those with suprasellar tumor extension. In a meta-analysis, DeKoltz et al. observed a statistically higher rate of visual improvement in the endoscopically treated patients when compared with the microsurgical group (71 vs. 56 %, respectively) [8]. This difference has not been substantiated by multiple other reports, which found comparable outcomes for both techniques [1, 8, 11, 13, 32].

These similar results are probably explained by the fact that the visual outcome primarily depends on the timing and decompression of the optic apparatus, rather than the extent of resection itself. The favorable visual results achieved by both techniques necessitates further study with precise pre- and post-operative visual evaluation in a large number of patients to uncover any statistical differences.

Quality of life (QOL)

When postoperative recovery is compared in patients who had endoscopic surgery subsequent to their initial microscopic sublabial transsphenoidal surgery, the endoscopic approach was associated with less post-operative pain, better airflow and a shorter hospital stay [40–42]. Notably, Lwu et al. found that when their patients were asked about their preference regarding an endoscopic or microscopic procedure for a future procedure, they favored the endoscopic option [42]. Current validated predictors of a better QOL are the extent of resection and the use of packing and nasal splints [43, 44], both of which are better managed via endoscopic surgery. Endoscopy appears to provide better post-operative olfactory function and endocrinological outcome, both of which play an important role in long-term QOL [45, 46].

Complications

CSF leaks

Since post-operative CSF leak was believed to be the “Achilles heel” of endoscopic endonasal surgery, abundant data is available on its comparison with that of the classic microscopic approach. Multiple systematic reviews compared the CSF leak rates associated with endoscopic surgery, finding a higher but statistically insignificant rate when compared to those of microscopic surgery [7, 10, 13]. The reported incidence of postoperative CSF rhinorrhea in microscopic surgery series usually ranges between 1 and 3 % [21, 47]. The greater visualization of the sella with the endoscope encourages more exposure, dissection and is likely to cause a higher rate of intraoperative CSF leakage, thereby counterbalancing the potential for improved reconstructive closure through endoscopy.

While reconstructive techniques during microscopic surgery are limited, successive improvements in endoscopic endonasal closure techniques, including the use of pedicled nasoseptal flap and multi-layer reconstruction protocols (e.g., gasket seal and bilayer button closure techniques), have improved postoperative CSF leak rates to less than 5 % in extended approaches to non-adenomatous lesions [48–50]. Pituitary adenoma surgery is classically associated with fewer post-operative leaks. The selective use of lumbar drainage for larger adenomas, together with administration of intrathecal fluorescein for identification of intraoperative CSF leaks, have further diminished the CSF leak rate to less than 1 % in the recent literature. These results are at least equivalent if not superior to the traditional microscopic outcomes [11, 12, 51–54].

Meningitis

Postoperative meningitis risk is broadly considered equivalent between endoscopic and microscopic surgery at around 1 % [2, 8, 10, 13, 46, 55]. A meta-analysis by Ammirati et al. found statistically less meningitis rates in endoscopic patients compared to microscopic: 1 vs. 2 % [7]. These results may be related to the better intra-operative identification and repair of CSF leaks offered by endoscopy [51].

Posterior pituitary dysfunction: diabetes insipidus

Disruption and manipulation of the posterior pituitary gland, pituitary stalk, or other fibers arising from the hypothalamus are frequently responsible for transient postoperative diabetes insipidus (DI) that might persist as a permanent deficiency. The majority of reports suggest that endoscopic resection results in less transient and postoperative permanent DI [7, 10–13]. The rate of transient DI ranges from 4–14 % for endoscopic resection, and 7.6–19.5 % for microscopic approaches. Similarly, for permanent DI, these rates range from 0–2 and 2–10 % for endoscopic and microscopic resection, respectively (Table 1). The risk for permanent DI with endoscopic procedures is less than 1 % in the recent literature [12, 55]. These results are based on retrospective review of nonrandomized studies. The enhanced visualization of the different structures, decreasing blind manoeuvers and inadvertent resection the posterior pituitary gland, plays an important role in improving the endoscopic results.

Table 1 Surgical results and complications from systematic reviews, meta-analysis, and single centers’ transition experiences, analyzing microscopic versus endoscopic pituitary surgery
Full size table

Anterior pituitary dysfunction

The reported rate of post-operative anterior pituitary dysfunction after microscopic endonasal surgery is around 3 % [9, 56]. Multiple reports suggest a lower rate of postoperative hypopituitarism linked to endoscopic surgery, though more data need to be gathered to reach sufficient statistical significance [7, 9, 10]. Recent results support the excellence of endoscopic techniques, with a reported anterior pituitary dysfunction rate of only 1.29 % in a large series of 1166 patients [55].

Visual complications

No statistically significant difference between the two techniques has been reported in the literature [7, 8, 10–12, 46, 57] for this parameter, although a trend of fewer visual complications is noticeable with endoscopic surgery (0.3–0.6 %) [7, 10, 55], compared to microscopic surgery 0.5–0.9 % [21, 23].

Vascular complications

Depending on the definition of a vascular complication, the rate might be discordant between the reviews. Still, these complications are sporadic in pituitary surgery, with rates less than 0.5 % [12, 56]. Even though some authors noticed an increase in vascular complications with endoscopic procedures [7], global consensus of the risk remains equivocal between the two techniques [2, 8–11, 13, 57–59]. Plausibly, endoscopic surgery provides better visualization laterally into the cavernous sinus, which encourages manipulation of its contents, leading to eventual vascular injury. This risk is balanced by the avoidance of any intrasellar blind instrument maneuvering performed during microsurgical approaches.

Rhinological complications

Multiple sino-nasal complications can shadow pituitary surgery. Sinusitis and nasal septal perforation are the most reported complications. Endoscopic surgery offers excellent results for postoperative sino-nasal QOL [43, 44], and these results are significantly superior to those achieved via microscopic surgery [8–10, 36, 40–42], which is more frequently associated with postoperative epistaxis, lip and nasal anesthesia, septum deviation, and postoperative synechiae [10]. The overall sino-nasal complication rate of microscopic surgery can reach 13 %, while only 1.2 % of patients developed such complications in the endoscopic surgery group [10]. There is evidence showing, however, that at 1 year after surgery, the results are equivalent [60]. The superior rhinological results achieved endoscopically are probably associated with the better visualization during all surgical steps, and minimizing any blunt or blind instrument manoeuvers. The close pre, intra, and post-operative collaboration with experienced otorhinolaryngologists also plays a crucial role in decreasing rhinological complications.

Economics

One of the old criticisms of endoscopic surgery was its economical weight. Current reports are converging to conclude that the overall cost per patient per procedure and global cost-effectiveness of endoscopic endonasal surgery is significantly more optimized when compared to the classic microscopic procedures [59, 61]. Although some authors report shorter operating time [11, 13], the use of two surgeons and the time it takes for the otolaryngologist to perform the approach may increase the operative time for the endoscopic approach. The endoscopic approach, however, is associated with shorter hospital stays with lower hospital resource consumption [11, 13, 57, 61].

The overall lower complication rate, extent of resection and adjunctive treatments (e.g., radiosurgery, medications, additional operations) and the less frequent need for short- and long -term endocrinologic replacement therapy are other factors to consider when evaluating the long-term cost effectiveness [8, 61]. Using a variety of economic models, a statistical difference favoring endoscopic surgery is noticeable [59, 61, 62] (Fig. 1). These results indicate that in large academic centers, endoscopic endonasal surgery is less costly 94–98 % of time [59].

Fig. 1
figure 1

Comparative Evaluation of Microscopic versus Endoscopic Cost-Effectiveness for a trans-sphenoidal procedure (US $)

Full size image

Surgeon skillset

Learning curve

Learning curves are an inevitable part of any new technique [1–3, 54, 63, 64]. In its infancy, endoscopic endonasal surgery was associated with mixed results, longer operating times and higher complications, particularly when compared to the good results achieved by the well-established standard microsurgical procedures. Currently, in specialized hands, endoscopic endonasal surgery results are equivalent or superior to microsurgery for almost all parasellar tumor extension, and equivalent to microsurgery for sellar pathology. Yet, some reluctance persists with its adoption, especially among microscopically trained neurosurgeons, as they are legitimately unwilling to accept the complications that might result as part of the learning curve.

Multiple experienced teams in microsurgery have now published their transition to endoscopy with unanimous results showing the safety of endoscopic endonasal surgery though the transition phase [1–3, 6]. However, comparison between results achieved sequentially must always be looked at skeptically due to the element of reporting bias. Interestingly, a recent report found similar surgical results when comparing the outcome of a less-experienced surgeon using a fully endoscopic technique to the outcome achieved by a very experienced surgeon using the microscopic transsphenoidal technique [65].

Beyond the sella: extended approaches

Extended microscopic approaches have been described by some authors. Even though some positive results were achieved in experienced hands [15–17], the degree of technical difficulty and the associated high morbidity rates prevented the spread of such techniques [15–17]. The concomitant development of standard and extended endoscopic approaches also prevented the wide development of such extended microscopic approaches.

Through the strong collaboration between neurosurgeons and otorhinolaryngologists and the modern improvement of endoscopic instruments and techniques, endoscopic endonasal surgery is succeeding in tackling an increasing array of pathologies ranging from simple CSF leaks to complex skull base tumors [66, 67]. Successive reports are progressively expanding the range of endoscopic surgery; from midline transclival approaches to brain stem tumors, including lateral approaches to the cervico-occipital junction and jugular foramen [66–68].

The continuous development of endonasal techniques also pushes the boundaries of other endoscopic-assisted cranial, skull base and neuro-endoscopic procedures. As a versatile technique, endonasal endoscopic surgery is an indispensable tool in the armamentarium of the contemporary skull base surgeon.

Conclusion

For many surgeons, the endoscopic transsphenoidal approach offers several advantages over the microscopic approach, particularly for removal of the extra- and parasellar tumor extension. Purely intrasellar adenomas can be removed, either using the microscope or the endoscope, with similar results.

Since there is a learning curve associated with the adoption of endoscopy, the selective use of the endoscope with only extrasellar tumors will prolong this learning curve. Therefore, most centers who use the endoscope use it exclusively for all endonasal tumor resections, and in doing so, have all abandoned the microscope.