Visual Disorders Outcome after Endoscopic Endonasal Trans-Sphenoidal Surgery of Pituitary Macroadenomas

Objective: Describe the outcome of visual disorders after endonasal 
transsphenoidal (EET) surgery of pituitary macroadenomas with preoperative chiasmal 
compression. Patients and Methods: From 2009 to 2013, 225 patients underwent 
EET surgery for sellar tumor. Among them, 168 suffered from visual disorders 
induced by chiasmatic compression, of which 88 met the inclusion criteria for 
this study. Average duration of follow up was 1.8 yrs ± 0.4 yrs. All patients 
had sellar MRI before surgery and yearly postoperatively, visual acuity (VA) 
and/or visual field (VF) measurements before, 3 months after surgery and yearly 
postoperatively. Results: All tumors were macroadenomas with a mean 
preoperative MRI-estimated volume of 10.0 {plus minus} 9.07 cm3. 
Preoperative VA was impaired with an average of 0.43 ± 0.13 LogMAR (left eye) 
and 0.36 ± 0.14 LogMAR (right eye) and VF was disturbed in 99% ± 2% of the 
cases with the most frequent disorder being bitemporal hemianopsia (38 
patients). Post-operatively, the mean residual MRI-estimated tumor volume was 
3.15 ± 3.71 cm3. Mean tumor volume reduction was 62% ± 9% (p < 0.001). 
Optic chiasmal compression resolved in 72% ± 10% of the cases. Visual function 
improved in 86% ± 7% of cases (p < 0.001). Conclusion: Endoscopic endonasal 
management of pituitary gland neoplasms is effective to reduce tumor volume. 
This technique achieved significant visual improvement in the majority of cases 
presenting with chiasmal compression syndrome.


Introduction
Pituitary adenomas accounts for about 10% of intracranial tumors among which 14% to 28% are non-functioning adenomas [1]- [3]. Such cases spontaneously grow up yielding to typical visual disorders related to optic chiasmal compression. Nowadays, the incidence of optic chiasmal compression is observed with MRI in 9% of the pituitary macroadenomas [4]. More rarely, involvement of the cavernous sinus leads to oculomotor symptoms. Such complications justify surgery to avoid visual degradation and to stop tumor growth progression. Endoscopic endonasal transsphenoidal (EET) surgery proved to be an effective procedure for such cases [2] [5] [6].
The aim of this retrospective study was to analyze visual outcome after EET surgery for pituitary macroadenomas.

Patients and Methods
From 2009 to 2013, 225 patients underwent surgical resection of pituitary tumors via endoscopic endonasal trans-sphenoidal approach. From this cohort, we retrospectively collected cases with visual impairment due to optic chiasmal compression by pituitary macroadenoma. Inclusion criteria were: 1) preoperative visual impairment (defined by visual loss and/or campimetric deficit and/or other symptoms like a diplopia); 2) pre and post operative MRI and visual status available; 3) histologically proven pituitary adenoma; 4) minimum of 3 months follow-up after surgery. Among 225 patients, 168 presented visual impairment from chiasm compression, and 88 cases met criteria for inclusion. 39 patients were females and 49 males (sex ratio 1.26). Mean age at surgery was 54.53 ± 3.01 years. All tumors were macroadenomas (n = 76; diameter ≥ 10 mm) or giant adenomas (n = 12; diameter ≥ 40 mm). The majority of tumors were gonadotropic (n = 72 patients). Baseline characteristics are summarized in Table 1.
All the patients had a preoperative MRI evaluation of the size of the pituitary adenoma and a CT scan to preoperatively study nasal fossa, sinus and skull base anatomy and for surgical navigation guidance. Tumor volume has been approximated based on MRI measurements, using an ellipsoid model (volume = 4/3 × π × 1/2 d 1 × 1/2 d 2 × 1/2 d 3 ) where d 1 , d 2 and d 3 are the maximum tumor diameter in the orthogonal spatial axes [7].
All the patients had a visual evaluation including measurement of visual acuity (VA) and/or visual field (VF) by Goldman or Humphrey perimeter. VA impairment was expressed in LogMAR (LogMAR 0 corresponding to a perfect vision and LogMAR 2.3 to a very low vision). All the patients underwent EET removal of the tumor with neuronavigation guidance. The surgery was carried out by a team comprising a neurosurgeon and an ENT surgeon, both experienced in "four hands" EE surgery. The surgical goals were to achieve maximum decompression of the optic chiasm, to remove as much as possible of the tumoral volume and to provide tissue for histological analysis.
Two major criteria were assessed: diminution of the MRI-estimated tumor volume, and visual outcome measured by VA and/or VF at the last follow-up visit.
The mean postoperative follow up was of 1.8 ± 0.4 years.

Statistical Analysis
Data were retrieved from the patient's hospital files, encoded in a Microsoft Excel spreadsheet and analyzed using the R statistical package. Means were compared before/after surgery using paired Student's test and proportions were compared using Mc Nemar test with Yates correction, with alpha 5%. Confidence intervals are 95%.
Reduction of the tumor was superior to 70% of its initial volume in 50% of the cases (n = 44); resection was near-complete (more than 90% reduction of the tumoral volume) in 23.8% of the cases (n = 21) (Figure 1 and Figure 2).

Visual Function
Prior to surgery, 4.5% of the patients (n = 4) presented a pituitary apoplexy and were not tested because of the surgical emergency of the case. Among the whole cohort, 6.8% of the patients (n = 6) had diplopia despite no visual defect or VF abnormality.
The mean preoperative VA impairment was 0.43 ± 0.13 LogMAR for the left eye and 0.36 ± 0.14 LogMAR   for the right eye. The mean postoperative VA impairment was of 0.16 ± 0.1 LogMAR for the left eye and of 0.16 ± 0.11 LogMAR for the right eye (Figure 3 and Figure 4). Mean VA impairment after surgery was 0.24 ± 0.13 Log-MAR for the left eye and 0.24 ± 0.12 LogMAR for the right eye. The difference before/after surgery was significant (p < 0.01).
If comparing pre and postoperative VA, data were available for 63 patients. There was a clinically significant improvement of the VA (average VA of both eyes improving by 0.2 LogMAR or more) in 47% of the cases (n = 30/63).
Assessment of VF 3 months after surgery was performed in 78 patients. It showed complete recovery in 15 cases, partial recovery in 27 cases and no improvement in 36 cases ( Table 3).

Discussion
EET surgery is a widely accepted procedure for the resection of micro, macro and giant pituitary adenoma [1] [5] [8]- [10]. It allows total resection in almost 20% of the cases in macroadenomas and giant adenom Apr;    [14]. Such encouraging results are explained by an improved panoramic visualization with EET, particularly for structures like the optic-carotid recess and the diaphragm sella. In case of visual impairment, EET is also indicated and yields good results [3]. In our article, reduction of the tumor was superior to 70% of its initial volume in 50% of the cases. There was a clinically significant improvement of the VA in 47% of the cases (n = 30/63). We observed 17% of complete recovery of the VA and 19.23% of complete recovery of the VF (n = 15/78).

Visual Tests
Between 2009 and 2013, 225 patients underwent EET resection of a sellar tumor at our institution. Among them 168 (74.6%) presented visual disorders, in agreement with results reported in the literature with rates ranging from 31% to 89% [1] [4] [6] [15]- [17]. In our series, 88 patients met inclusion criteria. Other patients were not eligible for various reasons (early lost of follow-up, unavailable MRI or visual examination data, other histological type of tumor). The functional results were better for VF (total recovery in 19.2% of the cases and partial recovery in 35%) than for VA (total recovery in 17% and improvement in 30% of the cases). This was noted in a series of juraska et al. [2]. On the other hand, Peter et al. [4] found better outcomes concerning VF than VA.
Cohen et al. [18] found an inverse correlation between the duration of visual symptoms and final outcome for both VF and VA.

MRI
In the literature, many authors recommended calculating tumor volume rather than diameters because it is a significant predictor of imaging outcome on long-term follow-up [13]. Hoffstetter et al. suggested a volume of ≥10 cm 3 as a modern definition of pituitary macroadenomas [19]. However, as noted by Egger et al. simple methods as geometric models like the ellipsoidal model provide a rough approximation of the tumor volume [7].
In our retrospective study, tumor volume was estimated using an ellipsoid model. We believe that with this model, volumes may be underestimated but the results were consistent with those found in the literature [20]. For Juraska et al. [2], volumetric analysis provides a more accurate measurement of the extent of the resection than ellipsoidal volume calculation. In our series, we approximated the tumoral volume by calculating ellipsoidal volumes because native MRI was not always at the radiologist disposal to calculate volume using dedicated software.
Postoperatively, the residual tumor volume was 3.15 ± 3.71 cm 3 (p < 0.01). As might be expected, the rate of residual tumor is higher in macro-adenoma (and giant adenoma), particularly in suprasellar and cavernous sinus spreading [2] [13]. Juraska et al. reported complete resection in 24.2% in their series of 76 giant and large macro-adenomas [2]. In our series, 23.8% of the patients had total or near total resection.

Limits of the Study
This is a retrospective study so we couldn't study the medical chart of the 225 operated on between 2009 and 2013 because of insufficient data. We decided to focus on pituitary adenomas excluding all other histological type of pituitary tumors because we don't know the evolution of all types of tumors.

Conclusion
Visual disorders are common complications of pituitary adenoma natural growth. In this series, we found that endoscopic endonasal transsphenoidal resection of macroadenoma with chiasm compression syndrome was reliable in improving visual impairment.

Conflict of Interest
The authors have no personal financial or institutional interest in any of the drugs, materials or devices described in this article.