A total of 52 patients were operated in the first 10 years and 158 in the next 18 years (Table 1). The age ranged from 15 to 72 years (Table 2) and included 95 males and 115 females (Table 3). 101 patients had tumors on the left and right side each and 8 patients had bilateral tumors (Table 4).

A large portion of patients presented with features of raised intracranial pressure- 81% in the first group and 68.9% in the second group. All 52 patients in the first group (100%) and 155 out of 158 patients in the second group (98%) presented with complete unilateral hearing loss ^{[3] [4]}. 27 out of 52 (52%) in the first group and 84 out of 158 (53%) had facial weakness ^{[4] [2]} present preoperatively. V^th nerve dysfunction tested by corneal reflex was found in 71% of first group patients and 64.5% of patients in second group. 8% in group 1 and 14% in group 2 had 6 ^th nerve involvement. 27% of group 1 and 16% of group 2 patients had lower cranial nerve involvement. Pyramidal weakness was found in 29% of first group and 16% of second group patients. 54% of patients in group 1 and 55% in group 2 presented with cerebellar signs (Table 5).

Tumor size was measured as the maximum size in any one axis of the extrameatal portion of the tumor. ^[4]

Tumors were divided into two groups based on size into two groups: Small, less than 3.0 cm and Large, more than 3.0cm. ^[3][5] Of the 52 tumors excised in the first group, 10 were small (19%) and 42 were large (81%). Of the 158 patients in second group, 49 were small (31%) and 109 were large tumors (69%).

Hydrocephalus was present preoperatively in 33 patients in the first group (63%) and 114 patients in the second group (72%). It was managed by direct surgery in 58% in group 1 and 83% in group 2 patients.

Surgery was performed with the patient in lateral position. ^[3] A classic retromastoid retrosigmoid ^[1-5][6-10] suboccipital ^{[3][5][9][10][11]} transmeatal ^[3] approach was used in all but three cases. Complete macroscopic excision using standard microsurgical ^{[3][7][9][11]} techniques was attempted. Partial excision was done in 3 patients in the first group and 6 in the second for elderly, poor general condition patients to allow amelioration of symptoms whilst reducing the risks of surgery.

Patients were assessed clinically by the chief operating neurosurgeon (Prof. A. Raja) in the immediate post-operative period and again at long term follow up ranging from 3 months to 20 years. Information recorded included anatomical facial nerve preservation, functional outcome of facial nerve function, total or subtotal excision of tumor, need for further procedures, other complications including lower cranial nerve involvement and peri-operative or post-operative death. ^[4] Facial nerve function was assessed using House and Brackman grading system. ^[12]The outcome of facial function was classified into Good outcome including patients who had a same or better facial functional status compared to pre-operative status and a Bad outcome, which consisted of patients who had worsening of facial nerve function on follow up 3 months after surgery. ^[2] (Table 6)

The clinical information used for this study was derived from notes of Prof. A. Raja contained in hospital files kept for all patients. No independent assessment of facial function was available which is possibly a confounding factor, but the same surgeon had examined all patients and should therefore be consistent throughout the series.

In order to assess the effect of surgeon’s experience on the results, cases were analyzed in two periods, 1980-1990 and 1991 – 2007. Fifty-two cases had been operated in the first period and one hundred and fifty eight in the second period. (Table 1)

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Anatomical preservation of facial nerve was assessed at the time of surgery. ^[4]It was possible to definitely look for and preserve the nerve in only 10 of 52 cases (19%) in the first group whereas in the latter group 128 of 158 cases (81%), anatomical preservation was achieved. There has been a significant improvement in the frequency of nerve preservation over the study period as depicted in Figure 1. In the first 5 years of study which included 20 patients, facial nerve preservation was possible in only 10 % cases. This went up to 25% in the next 5 years and 68% in the subsequent 5 years. In the last 10 years of study, which included the largest group of 108 patients, anatomical preservation achieved has been 100%.

These results demonstrate an improvement in facial nerve preservation with increasing experience of the surgeon ^[4] irrespective of the size of tumors, which have been consistently large throughout the series.

In the first group of 10 patients with anatomical preservation, only 2 patients had an actual improvement in facial nerve function (based on House & Brackman grading^[12]). 2 patients continued to have function similar to pre-operative status and 6 patients worsened post-operatively.

In the second group of 128 patients with anatomical preservation, 35 patients (27%) improved in facial function, 71 patients (55%) continued to have same function as pre-operative and 22 patients (17%) had worse function post-operatively. (Table 7)

The patients who had not worsened and were better or remained the same post-operatively were classified as Good outcome and the patients who had worsened post-operatively in their facial function were classified as B ad outcome. In the first group, 4 patients (40%) had a Good outcome and 6 patients (60%) had a Bad outcome. Where as in the second group, 106 patients (82.8%) had a Good outcome compared to 22 patients (17.2%) having a Bad outcome. (Table 6)

The post-operative neurological deficits, apart from facial nerve, are recorded in Table 8. The most striking feature is continued eight nerve deficit in most of the patients. All other deficits have shown marked improvement.

The complications occurred in our patient groups are recorded in Table 9. The main complications have been bleeding ^[1][3], CSF leak ^[1][3][4][8] and meningitis ^[1][3] post-operatively.

7 patients (13%) died in the first group where as mortality was significantly lower in the second group where only 4 patients (2.5%) died. The reasons for mortality have been enumerated in Table 10.

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

The facial nerve was anatomically preserved in 19% in the first group and 81% in the second group. The use of intraoperative facial nerve monitoring ^[2][3][11] was done in a few cases but has not been continued subsequently. ^[1]The drilling ^[1][6] of internal auditory meatus at the outset and visualizing the facial nerve in the meatus and then going for complete excision of the tumor has led to increasing rates of facial nerve preservation in all of the last 108 cases. ^[3][9]We feel that early drilling of internal auditory meatus and facial nerve visualization has been the single most important factor in preservation of facial nerve. The rates of preservation in our study have been similar to studies done by Samii et al (1997)^[11] and Yamakami et al (2008)^[3], where intraoperative facial monitoring has been done and attributed as the cause of better rate of preservation rather than drilling of internal auditory meatus.

Good Facial nerve functional outcome at 3 months had improved from 40% in the first group to 82.2% in the second group. This is comparable to major studies by Samii et al ^[11] and Elsmore et al^[4]. But, in their studies the majority of tumors were small and medium sized whereas in our study, majority of tumors were large- 81% in first and 69% in second group. Elsmore et al ^[4] had concluded that rates of functional preservation in large tumors were poor (13.5%), even in cases where they were able to anatomically preserve the nerve.

In our study, we have achieved good functional outcome even with two thirds of our tumors being large. So we believe that tumor size is not a reliable predictor of facial nerve functional outcome. There may be other factors which may influence functional outcome. In our experience we noticed that consistency of the tumor may be one such factor. Soft and cystic tumors, which had only displaced the nerve, have a better outcome after tumor excision whereas hard tumors compress the nerve and cause permanent damage somehow and facial function does not seem to improve inspite of anatomical preservation. Figures 2 – 5 show a profile of patients along with their pre- and post- operative scans, all showing good facial nerve functional preservation.

The chance of preserving useful hearing in large tumors is minimal^[3]. We had most patients presenting with complete hearing loss and therefore could not be benefitted by the procedure. Three patients had residual hearing and eighth nerve was preserved intraoperative. Out of these, only one patient improved and had useful hearing, one patient had some hearing with poor discrimination and the third had complete hearing loss.

The mortality has come down from 13.5% in the first group to 2.5% in the second group. This is comparable to large studies be Samii et al ^[11] and Elsmore et al^[4], where it has been quoted between 1% to 3%. We believe that mortality can be reduced further if patients come early to the hospital and not in moribund state as so many in India do.