An Analysis of Retinal Nerve Fiber Layer Thickness before and after Pituitary Adenoma Surgery and its Correlation with Visual Acuity
Keywords: Optical coherence tomography, pituitary adenoma, retinal nerve fiber layer thicknessKey Messages: RNFLT can be used as prognosticating factor in pituitary adenoma surgery for improvement in vision, its regular pre-operative evaluation can be done for better councelling of patient.
Pituitary adenomas comprise approximately 10% of all intracranial tumors, and 50–80% of pituitary tumors are pituitary adenomas which are benign lesions; pituitary carcinoma are rare. Symptoms related to pituitary tumors are due to mass affect in nonsecreting tumors if they are sufficiently large; on the other hand, secreting tumorscause clinical syndromes due to abnormal hormonal secretion and their effect on different target organs. The most common neuro-ophthalmological syndrome associated with pituitary adenomas is caused by compression of the central part of the optic chiasm, which produces the classic bitemporal hemianopia in the visual field.
Vision loss in pituitary tumor occurs as a consequence of suprasellar growth and compression of anterior visual pathways. The superior temporal quadrants tend to be affected first, followed by the inferior temporal quadrants. Subtle changes in vision can be picked up in perimetry and visual acuity tests.
It has been seen that following surgery of pituitary macroadenoma, total recovery of normal vision occurs in 35% of the patients, improvement of vision in 60%, and in the rest there is no change in vision. Factors responsible for recovery of vision include the degree of optic apparatus decompression during surgery, preoperative vision loss, and duration of symptoms. Despite having good tumor decompression, some patients do not show improvement in vision. This could be due to changes in the proximal part of visual pathway. These changes could be in rods, cones, optic disc, or retinal nerve fiber layer.
Retinal nerve fiber layer thickness (RNFLT) undergoes retrograde degeneration following compression of optic apparatus by the pituitary tumor. RNFL is the ninth layer of the 10-layer retina whose thickness can be measured by using optic coherence tomography (OCT). OCT is a valuable tool for diagnosis and prognosis by estimating axonal loss. Comparing the changes in RNFLT after surgery and its correlation with visual outcome may help to define its prognostic value.
With this background, we planned a study to evaluate RNFLT before and after pituitary surgery and its correlation with visual acuity. Moreover, fundus was also evaluated to rule out the grade of damage to optic disc and to determine whether in-vivo RNFLT measurement predicts the visual recovery of such patients.
We conducted a study on 20 patients (40 eyes) who were diagnosed with pituitary adenoma at GB Pant Hospital, New Delhi from February 2015 to October 2016. Preoperative RNFLT was measured using OCT, and postoperative measurement was done after 1 and 3 months. Four-quadrant mean of RNFLT was calculated. Both transnasal and transphenoidal or transcranial approach was done according to patient indications. Inclusion criteria had all patients with pituitary adenoma with change in their vision; aged >18 years and <50 years; patients whose tumor was totally removed or optic apparatus was adequately decompressed. Patients who had residual tumor on postoperative CT scan were excluded from the study. All patients who had previous ophthalmological disease were also excluded from the study.
After a full refractive correction, the visual acuity of each patient was tested using a retro illuminated Snellen's chart at a distance of 6 m. Snellen's visual acuity was then converted to get decimal acuity for the ease of statistical analysis. Visual acuity was measured preoperatively and 7 days, 1 month, and 3 months after the procedure.
RNFLT was calculated using spectral-domain OCT Optovue (Heidelberg Engineering, Heidelberg, Germany (RT 100 version 5.1); [Figure 1]. We obtained RNFLT using RNFL 3.45 scanning protocol. The RNFL scan pattern completes four circular scans in 0.16 s at a diameter of 3.45 mm, targeted around the optic nerve head. Preoperative measurement was done followed by postoperative measurement after one and three months. Four-quadrant mean of RNFLT was calculated [Figure 2]. These scans are averaged and the result is presented within the normative range parameters. Different studies have been conducted to know the normal values of RNFLT according to age, race, and gender. No statistical difference was found on the basis of sex, but different studies have shown different results in age and race. We calculated the average from different studies  reported in Indian population and found that mean ± SD was 95 μm ± 10 μm; thus, we considered normal RNFLT to range from 95 μm ± 10 μm in 18–50-year age group. The maximum value for RNFLT can reach up to 189 μm, but it occurs in few people.
The fundus of each patient was examined using the conventional direct ophthalmoscope. Changes in optic disc were looked for:
Grading of pallor was as follows:
+: Mild disc pallor
++: Moderate disc pallor
+++: Severe disc pallor.
Twenty patients (40 eyes) were studied with ages ranging from 18 to 50 years, out of which 11 were males and 9 were females. Eighteen patients presented with visual symptoms for less than one year, only two patients had symptoms for more than one year. According to Hardy Wilson grading, 7 (35%) patients were found in the grade IIA, that is, enlarging sella and extending into suprasellar cistern with compression of chiasma, followed by 5 (25%) in the grade IIB, and 3 (15%) in the grade IIC. Only 1 (5%) patient each was found in grades IIIA, IIIB, IIIC, IIID, IVB [Table 1].
Preoperatively 19 (47.5%) eyes had RNFLT <85 μm, 13 (32.5%) eyes had RNFLT between 85 μm and 105 μm, whereas 8 (20%) had thickness >105 μm. Postoperatively at one month 16 (40%) eyes were below normal thickness, 15 (37.5%) eyes had normal thickness, and 9 (22.5%) eyes had more than normal thickness. Similarly, postoperatively at three months, 16 (40%) eyes had thickness below normal, 16 (40%) eyes had thickness in the normal range, and 8 (20%) eyes had thickness greater than normal [Table 2].
The mean preoperative RNFLT was 89.22 ± 18.33 μm and postoperative RNFLT at 1 month was 89.35 ± 18.21 μm. This difference before and after the surgery was found to be statistically nonsignificant (P = 0.674). The mean postoperative RNFLT at 3 months was 89.46 ± 18.76 μm. This difference before and after the surgery was found to be statistically nonsignificant (P = 0.470). Thus, no significant changes occur in RNFLT after surgery both at one and three months of follow-up [Table 3].
Preoperative visual acuity for majority, i.e., 15 (37.5%) eyes, were found in the range of no perception of light (PL-) to 6/60 followed by 14 (35%) eyes in the range 6/12 to 6/6. Eleven (27.5%) eyes were found in the range of 6/36 to 6/18. Postoperative visual acuity on day 7 for majority, i.e., 15 (37.5%) eyes were found in the range of 6/12 to 6/6 followed by 13 (32.5%) eyes in the range 6/36 to 6/18. Twelve (30%) eyes were found in the range of no perception of light to 6/60. Postoperative visual acuity at 1 month for majority, i.e., 16 (40%) eyes were found in the range of 6/12 to 6/6, followed by 13 (32.5%) in the range 6/36 to 6/18. Eleven (27.5%) eyes were found in the range of no perception of light (PL-) to 6/60. Postoperative visual acuity at 3 months for majority, i.e., 20 (50%) eyes were found in the range of 6/12 to 6/6, followed by 12 (30%) eyes in the range 6/36 to 6/18 and 8 (20%) eyes were in the range of perception of light (PL-) to 6/60 [Table 4].
Preoperative mean visual acuity in decimal fraction was 0.31 ± 0.29, which improved to 0.38 ± 0.34 after the seventh postoperative day, 0.47 ± 0.39 after one month and 0.53 ± 0.43 after 3 months. This difference of mean pre and postoperative at 7th day, 1 month, and 3 months was found to be statistically significant (P<.05). This shows there was improvement of visual acuity after surgery and changes were noticed just after 7 days, which continued to improve till 3 months of follow-up [Table 5].
On correlating RNFLT with visual acuity after 3 months of surgery, we found that 16 eyes had RNFLT <85 μm, of which eight eyes had VA between 6/60-PL(-), 6 eyes had VA 6/36-6/18, and two eyes had VA 6/6-6/12. Similarly, 16 eyes had RNFLT between 85 and 105 μm, of these 10 eyes had VA 6/6-6-12 and 6 eyes had VA 6/18-6/36. Lastly, 8 eyes had RNFLT >105 μm and all eyes had VA 6/6-6-12. Thus, we found that there was improvement in vision in patients who had RNFLT in the normal range (mean ± SD) or greater, and that there was no or little improvement in vision among patients who had preoperatively thinned-out RNFL [Figure 3].
Preoperative optic disc findings in majority, i.e., 14 (35%) eyes were normal, whereas in 11 (27.5%) eyes mild pallor (+) was noted, nine (22.5%) eyes had moderate pallor (++), and 6 (15%) eyes had severe pallor (+++). Optic disc examination findings after 1 month of surgery in 16 (40%) eyes were normal whereas 9 (22.5%) eyes had mild pallor (+), 9 (22.5%) eyes had moderate pallor (++), and 6 (15%) eyes had severe disc pallor (+++). Optic disc examination finding after 3 months of surgery was similar as that seen after 1 month. Thus, no change was seen in optic disc with severe pallor and moderate pallor, only two patients showed improvement from mild disc pallor to normal optic disc. The Chi-square test was applied to pre and postoperative optic disc color, and P value was found to be nonsignificant (P > 0.05) [Table 6].
Studies have been done to demonstrate the usefulness of OCT in estimating RNFL thickness in cases of pituitary adenoma. Normally, thinner RNFL measurements are associated with older age and increasing myopia. Caucasians tend to have thinner RNFL values when compared with Hispanics and Asians. The thickest RNFL measurements were found in the inferior quadrant, followed by the superior, nasal, and temporal quadrants.
OCT is a noninvasive technique that allows cross-sectional imaging of the retina and quantifies the thickness of the RNFL around the optic nerve head. OCT can help in the diagnosis of adenomas in equivocal situations where there is only subtle optic nerve pallor in the setting of nonspecific visual complaints or visual field defects. The degree of RNFL thickness reduction has been shown to correlate with that of visual field defects.
In the present study, it was observed that mean RNFLT thickness preoperatively was 89.22 ± 18.33 μm, postoperatively after one month was 89.35 ± 18.21 μm, and postoperative after 3 months was 89.46 ± 18.76 μm. There was no change in RNFLT pre and postoperatively. The comparison between pre and postoperative RNFLT at 1 and 3 months was found to be statistically insignificant.
Visual improvement after trans-sphenoidal adenomectomy is very rapid. In a recent study, Kerrison et al. suggested that visual recovery occurs in at least three phases – an early fast phase, an early slow phase, and a late phase. Improvement in vision may take place immediately after decompression. Visual evoked potential have been documented to improve within 10 min of decompression. It has been postulated that the initial improvement is the result of the removal of a physiologic conduction block., Further improvement during a stage of delayed recovery is thought to be the result of re-myelination of the decompressed optic pathways.
In our study, we found that there was improvement in visual acuity postoperatively. Preoperatively, 14 (35%) eyes had vision 6/12-6/6; 1 month postoperatively, 16 (37.5%) eyes had vision 6/12-6/6; and 3 months postoperatively, 20 (50%) eyes had vision 6/12-6/6. Preoperative mean visual acuity in decimal fraction was 0.31 ± 0.29; 7 days postoperatively was 0.38 ± 0.34; 1 month postoperatively was 0.47 ± 0.39; 3 months postoperatively was 0.53 ± 0.43. This difference both before and after surgery was statistically significant (P < 0.05). None of the patients had deterioration of vision after surgery. The results of Dhasmana et al. are consistent with the present study showing that visual acuity improved in 17 (47.2%) eyes and remained unchanged in 19 (52.8%) eyes.
Dekkers et al. in their study found that visual acuity improved significantly within 3 months after trans-sphenoidal surgery and that it progressively improves after the first year
It was also found that improvement in visual acuity occurs postoperatively only when preoperative RNFLT was greater than 85 μm; for patients with RNFLT below 85 μm did not show any improvement of vision after surgery.
A study by Danesh-Meyer et al. (2015) showed similar results as our study. Helen (2008) observed that patients with measurable RNFL loss at the time of surgery for chiasmal compressive lesions had less recovery of VA or VF after surgery.
Gnanalingham et al. reported that recovery of visual field deficit takes place over many years after trans-sphenoidal approach for pituitary macroadenoma surgery, with recovery of visual field in 95% of the patients. According to the study, postoperative visual field defect determines the extent of visual field recovery.
Quio et al. (2015) conducted a study to determine RNFL changes after trans-sphenoidal approach and patients with the transcranial approach, concluding that mean RNFLT decreased in patients who underwent transcranial surgery and no changes were noted in trans-sphenoidal approach. Jacob et al. (2009) found out that RNFL thinning measured by OCT puts the patient at decreased chance of recovery postoperatively after treatment of pituitary adenomas compressing the anterior visual pathways.
Ferreri evaluated RNFLT using OCT 3 and found out that temporal RNFL show a reduced thickness in pituitary adenoma patient; however, no difference was found in other quadrants between normal participants and adenoma patients and between patients who were pharmacologically treated and patients who underwent surgical adenomectomy.
Moon et al. reported significant improvement in visual field post optic chiasmal decompression, and RNFLT was reduced 3 months post surgery; however, there was a significant improvement after 6 months of surgery.
In our study, we also concluded that optic disc changes are permanent and do not revert if they exhibit moderate to severe pallor; only mild pallor has a chance to revert to normal if surgery is done as early as possible. Optic disc pallor is also associated with visual acuity; usually a severe pallor optic disc has poor visual acuity, and there is no improvement in eyes even after surgery. Thus, timing of surgery is very important, and for safe vision it should be done before optic disc atrophy sets in due to chronic compression of chiasma by pituitary tumor.
In 1985, Jain et al. reported that they had studied five patients with pituitary adenoma, evaluating, both pre and postoperatively, the peripapillary nerve fibres by red free fundus examination to correlate the preoperative assessment of the optic atrophy with the postoperative outcome. They evaluated visual acuity, fundus, and visual field anatomy in Bjerrum's screen. They concluded that this method of evaluation, though subjective, showed that low grades optic atrophy preoperatively are associated with good visual outcome. Preoperatively, visual loss is disproportionate to the degree of optic atrophy, and final improvement in visual acuity and visual field can be predicted on the basis of optic atrophy score. They also concluded that a good capillary count indicated a good perfusion of the optic nerve head, and there was a reliable indicator of visual prognosis.
In our study, we have found that optic disc and RNFLT is an important prognosticating marker for visual recovery in pituitary adenoma patients after surgery. Great concern should be taken while evaluating patients' fundus, and even subtle changes in optic disc should prompt surgeons for early surgery, as it was found that only mild pallor can revert to normal and moderate and severe pallor do not revert to normal, causing deterioration of vision. Once optic atrophy sets in, patients become permanently blind.
RNFLT provides a quantitative measure of the amount of viable axonal tissue. It is a novel preoperative marker that aids in predicting the degree of visual recovery in patients with chiasmal compression after a decompression procedure. The benefit in having such a clinical biomarker include being rapid, noninvasive, and convenient to measure. Its usefulness in planning of management strategies including the timing of surgery and counselling patients on prognosis. However, further studies are required to corroborate these preliminary findings in larger studies and in wider range of underlying diseases.
This work is dedicated to my patients without their cooperation, study would have not possible.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]